Der8auer successfully delids Intel's 12-18 core Skylake-X CPUs

[AlienALX]

What you quoted there is my opinion, nothing more. Please tell me where I said it was 100% proven fact? However, there is a lot of evidence to back up my opinion. Where as his is 50% out at least because AMD.

As I said to Kaap, please tell me what you think would happen to a 6700k if it launched and ran 18c cooler? exactly what would people do to it?

 

[AngryGoldfish]

What you quoted there is my opinion, nothing more. Please tell me where I said it was 100% proven fact? However, there is a lot of evidence to back up my opinion. Where as his is 50% out at least because AMD.

As I said to Kaap, please tell me what you think would happen to a 6700k if it launched and ran 18c cooler? exactly what would people do to it?

I briefly read through Der8auer's post and found the majority of it to be observations based on personal testing and knowledge. At no point did I notice him doing anything more than you in your post. In fact, here is excerpt from his conclusion:

Skylake works great with normal paste so I see no reason why Intel should/would change anything.

Objectively, Skylake does work great. Temperatures are not through the roof. A 4.6Ghz overclock is very attainable with a good air cooler. It could work better, but he doesn't state that as a fact himself. He simply shows his evidence and uses that to do the talking. Again, that's all I've read.

As for the part where he says "I see no reason", focus on the 'I'. He's claiming to have an opinion based on findings. That's no different than what you or I do.

I 100% promise you...

vs

I see no reason...

They both suggest confidence in what's being said.

 

[AlienALX]

I think you missed my two attempts at humour.

1. Twatism.
2. I 100% promise you.

 

[Kaapstad]

What you quoted there is my opinion, nothing more. Please tell me where I said it was 100% proven fact? However, there is a lot of evidence to back up my opinion. Where as his is 50% out at least because AMD.

As I said to Kaap, please tell me what you think would happen to a 6700k if it launched and ran 18c cooler? exactly what would people do to it?

Perhaps you could do a load of testing like Der8auer has and post your findings until then you have nothing but your opinion based on nothing.

IIRC too in Der8auer's article he has posted photographic evidence of where solder causes cracking.

What have you got - nothing.

 

[AngryGoldfish]

I think you missed my two attempts at humour.

1. Twatism.
2. I 100% promise you.

Did I?

But in all seriousness:

 

[Bartacus]

He also said in his article that he was focused on extreme thermal cycling, with sub-zero temperatures, which is how he got his results. The average joe (and most OCers in fact) do not dabble in sub-zero cooling, therefore his article is worthless in reference to Intel / AMD shipping soldered CPUs.

If it was geared toward sub-zero OCers, then fine. But to do that kind of testing and then claim "soldered CPUs are bad and will eventually crack" is B.S. Stupid article is stupid.

 

[AngryGoldfish]

He also said in his article that he was focused on extreme thermal cycling, with sub-zero temperatures, which is how he got his results. The average joe (and most OCers in fact) do not dabble in sub-zero cooling, therefore his article is worthless in reference to Intel / AMD shipping soldered CPUs.

If it was geared toward sub-zero OCers, then fine. But to do that kind of testing and then claim "soldered CPUs are bad and will eventually crack" is B.S. Stupid article is stupid.

But as far as I can see, he didn't say that.

Micro cracks in solder preforms can damage the CPU permanently after a certain amount of thermal cycles and time.

That's nowhere near as definitive and dismissive as your paraphrasing. In fact, the latter part is highly vague and inconclusive.

I don't particularly like Der8auer's way of doing things either, but I think some of you are misinterpreting his comments and unfairly giving him a hard time.

 

[AlienALX]

Perhaps you could do a load of testing like Der8auer has and post your findings until then you have nothing but your opinion based on nothing.

IIRC too in Der8auer's article he has posted photographic evidence of where solder causes cracking.

What have you got - nothing.

No, just over 30 years of experience with electronics and stuff. And soldering of course. I spent the best part of five years soldering for about 8 hours a day.

I've seen you argue with people before. You take one thing and then hold onto it like a dog with a bone, completely ignoring all of the facts around you. So, given that is the case answer my question.

Why is it that AMD have soldered CPUs ever since the first dual core and you've never had issues.

There. One question.

As for my experience with electronics?

https://forum.overclock3d.net/showthread.php?t=77462&page=8

There you go. Make sure you read it all of course, especially the bit where I -

1. Took a power supply apart and
2. Completely rewired it from the inside out.

The stereo has been on for 24 hours a day since and is still running fine.

 

[Kaapstad]

No, just over 30 years of experience with electronics and stuff. And soldering of course. I spent the best part of five years soldering for about 8 hours a day.

I've seen you argue with people before. You take one thing and then hold onto it like a dog with a bone, completely ignoring all of the facts around you. So, given that is the case answer my question.

Why is it that AMD have soldered CPUs ever since the first dual core and you've never had issues.

There. One question.

As for my experience with electronics?

https://forum.overclock3d.net/showthread.php?t=77462&page=8

There you go. Make sure you read it all of course, especially the bit where I -

1. Took a power supply apart and
2. Completely rewired it from the inside out.

The stereo has been on for 24 hours a day since and is still running fine.

I think you need to post some evidence first before this debate can go forward, to say you have done something for the last 30 years counts for squat without proof to support your argument.

 

[AlienALX]

I think you need to post some evidence first before this debate can go forward, to say you have done something for the last 30 years counts for squat without proof to support your argument.

See? exactly as I said. Post one thing, stick to it like a dog with a bone. I will ask you again.

Why is it that AMD have soldered CPUs ever since the first dual core and you've never had issues.

If you don't want to answer my question then please, don't reply to me.

Oh as for the proof? how many Sandybridge CPUs have you seen that idle at abnormal temps and shut the rig down because the solder has failed - that is my proof. Intel soldered CPUs for years and there has never been an issue with it, nor a "solder gate".

I don't particularly like Der8auer's way of doing things either, but I think some of you are misinterpreting his comments and unfairly giving him a hard time.

Then maybe he should stop going around posting his opinions like he knows it all.

https://www.reddit.com/r/intel/comments/6md8n6/der8aur_why_doesnt_intel_solder_cpus_its_not_just/

His opinion was fine until now. Right up until AMD come along and solder Ryzen, which is 14nm. Especially as Intel used the excuse in the past that Haswell could not be soldered because of the nm, and how the dies themselves would crack if they tried to solder them.

Which was obviously BS, because my Broadwell E CPU is 14nm and is soldered.

Now they are saying the solder cracks. What, on a £5000 CPU? (referring to the top of the range Xeon here).

That article does not make sense. It says Intel say that they can not solder certain types of die because of cracking (whether that be the die or the solder). However, they trip themselves up by soldering the same tech only as a Xeon.

Some one on Reddit had a good point, they are probably just doing it because they are cheap. That really could be all there is to it.

 

[Kaapstad]

See? exactly as I said. Post one thing, stick to it like a dog with a bone. I will ask you again.

Why is it that AMD have soldered CPUs ever since the first dual core and you've never had issues.

If you don't want to answer my question then please, don't reply to me.

Oh as for the proof? how many Sandybridge CPUs have you seen that idle at abnormal temps and shut the rig down because the solder has failed - that is my proof. Intel soldered CPUs for years and there has never been an issue with it, nor a "solder gate".



Then maybe he should stop going around posting his opinions like he knows it all.

https://www.reddit.com/r/intel/comments/6md8n6/der8aur_why_doesnt_intel_solder_cpus_its_not_just/

His opinion was fine until now. Right up until AMD come along and solder Ryzen, which is 14nm. Especially as Intel used the excuse in the past that Haswell could not be soldered because of the nm, and how the dies themselves would crack if they tried to solder them.

Which was obviously BS, because my Broadwell E CPU is 14nm and is soldered.

Now they are saying the solder cracks. What, on a £5000 CPU? (referring to the top of the range Xeon here).

That article does not make sense. It says Intel say that they can not solder certain types of die because of cracking (whether that be the die or the solder). However, they trip themselves up by soldering the same tech only as a Xeon.

Some one on Reddit had a good point, they are probably just doing it because they are cheap. That really could be all there is to it.

It does not matter how much you try you can not shout people down by repeating the same old stuff without proof and you have got none.

At least Der8auer produced some evidence and is not trying to shout down everyone who disagrees with him.

 

[AlienALX]

It does not matter how much you try you can not shout people down by repeating the same old stuff without proof and you have got none.

At least Der8auer produced some evidence and is not trying to shout down everyone who disagrees with him.

Why is it that AMD have soldered CPUs ever since the first dual core and you've never had issues.

Here we go, robot mode engaged.

Edit. Last I am going to say in this thread. In the Reddit thread Der8auer was asked the same question. Like you he either could not, or did not want to answer it. Probably because he was fed a load of BS by Intel and then AMD come along and quickly cut down that BS.

So yes, in a way a trick question. You can't answer it, can you? so there goes *your* proof.

 

[AngryGoldfish]

Then maybe he should stop going around posting his opinions like he knows it all.

https://www.reddit.com/r/intel/comments/6md8n6/der8aur_why_doesnt_intel_solder_cpus_its_not_just/

His opinion was fine until now. Right up until AMD come along and solder Ryzen, which is 14nm. Especially as Intel used the excuse in the past that Haswell could not be soldered because of the nm, and how the dies themselves would crack if they tried to solder them.

Which was obviously BS, because my Broadwell E CPU is 14nm and is soldered.

Now they are saying the solder cracks. What, on a £5000 CPU? (referring to the top of the range Xeon here).

That article does not make sense. It says Intel say that they can not solder certain types of die because of cracking (whether that be the die or the solder). However, they trip themselves up by soldering the same tech only as a Xeon.

Some one on Reddit had a good point, they are probably just doing it because they are cheap. That really could be all there is to it.

I can't find a post by Der8auer's in that Reddit thread. Are you linking just 'cause it's interesting (which it is) and correlates a lot with you're saying?

This post was particularly interesting:

Ryzen chips are indeed soldered and have no big thermal issues (although architecture and process node are very different, so not directly comparable)

The fact that Intel solder their Xeon chips is highly telling. But like I said at the beginning of this thread, there has to be a sound engineering reason for not soldering their CPUs. I struggle to accept the argument of 'Intel hates overclocking.' While there is evidence pointing to that, there's evidence pointing to the contrary. The argument of 'Intel are just lazy' doesn't make any sense. If they really were lazy, they wouldn't be so fired up to compete against Ryzen. 'Intel are cheapskates'. That one makes more sense. They have taken CPU's that once had 44 PCI-e lanes and then chopped them down, all while the competition offer high PCI-e lanes across the range. To me, 'Intel are cheapskates' and 'Intel have sound engineering reasons' are my two go-to for explaining this debacle.

 

[Kaapstad]

Why is it that AMD have soldered CPUs ever since the first dual core and you've never had issues.

Here we go, robot mode engaged.

Edit. Last I am going to say in this thread. In the Reddit thread Der8auer was asked the same question. Like you he either could not, or did not want to answer it. Probably because he was fed a load of BS by Intel and then AMD come along and quickly cut down that BS.

So yes, in a way a trick question. You can't answer it, can you? so there goes *your* proof.

You still have provided no proof.

Der8auer 1 - 0 AlienALX

 

[NeverBackDown]

Well, considering its hard to find a CPU that is soldered. That could be difficult to answer right now

Personally I wouldn't mind the paste if only they used a decent quality one. Would love to see the day they went into partnership with a reputable manufacturer so we got to see MX-4 being used, or something from Grizzly etc.

Every AMD CPU is soldered from recent memory and I am sure even before the Phenom's. They are fine.

Old Intel chips were soldered. They all still run. I have multiple CPUs in old machines, they still work. Temps aren't an issue on any of them as far as I can tell because they don't auto shut off. No point in trying to get temp software on them since they are so slow. But the point is, soldered is better and will far outlast any sane usage from a CPU.

 

[AngryGoldfish]

You still have provided no proof.

Der8auer 1 - 0 AlienALX

Now that was petty. You're ruining whatever argument you had so it might be better if we let this one go. NBD said it best.

Every AMD CPU is soldered from recent memory and I am sure even before the Phenom's. They are fine.

Old Intel chips were soldered. They all still run. I have multiple CPUs in old machines, they still work. Temps aren't an issue on any of them as far as I can tell because they don't auto shut off. No point in trying to get temp software on them since they are so slow. But the point is, soldered is better and will far outlast any sane usage from a CPU.

This.

/endofthread

 

[Kaapstad]

Now that was petty. You're ruining whatever argument you had so it might be better if we let this one go. NBD said it best.



This.

/endofthread

The point is Der8auer has put plenty of info on the table, AlienALX has offered nothing but his opinion which he trying to use to beat out what anyone else has to offer.

As to whether anyone has seen a faulty CPU where the solder has failed, this is very difficult to spot as the average user does not have the equipment to check.

 

[AlienALX]

The point is Der8auer has put plenty of info on the table, AlienALX has offered nothing but his opinion which he trying to use to beat out what anyone else has to offer.

As to whether anyone has seen a faulty CPU where the solder has failed, this is very difficult to spot as the average user does not have the equipment to check.

I've put forward more facts than you have. Far more.

Nice swerve in the ending sentence though. Yes, trying to get out of it by saying that nobody has the equipment to detect it is a good dodge.

The fact is that you can not dispute what I am saying. Not with any facts. And copying and pasting some one else's opinion does not get you off the hook either.

I'm telling you that I have had 30 years of electronics experience. At least that. I built my ZX80 with my uncle when I was 7, and was taking apart electronic items long before that. My shed was full of them.

So I think I know a wee bit more about solder, its properties and how it reacts than you do. AMD obviously do too. And Intel, when it suits them.

You've put forward your "facts" and I have disputed them. I asked you a simple question. Not even more than one line. you can not answer it because you do not know the answer, do you?

Therefore there is no argument. I politely repeated my question three times, each time you chose to ignore it and then after you had some time to mull it over you say "Nobody has the equipment to check".

Let me reiterate for you, just in case you do not understand electronics (which I am slowly beginning to think you know naff all) soldering a CPU does not create an electrical connection. It's simply flux and tin. If it cracks on the die and cracks enough to cause a problem (like if the flux has perished, as it does and this is why dry joints occur) then you would see an instant rocket in temps. Instant. Once that connection is gone your CPU would simply overheat.

And I have seen no cases of that ever with a soldered CPU. I have been on the internet since 1997 and I have been buying and reading computer mags since I was about 8, and not once have I ever seen a feature about solder on a CPU failing, or for that matter even being a problem. Ever. Not once.

The point is Der8auer has put plenty of info on the table, AlienALX has offered nothing but his opinion which he trying to use to beat out what anyone else has to offer.

How can you even so much as ponder starting a debate over something using some one else as cover? if you don't know what you are talking about you're better off staying out of it, no?

Go and ask him (seeing as you are using him as your go to) why AMD can solder Ryzen. Not that I trust him of course, given he looks about half my age. And when he gives you an answer bring it here. Let's see if he can explain it, because you obviously can't.

 

[WYP]

The simple fact here guys is that the majority of Intel's CPUs are not intended to be overclocked, so even if Intel changed from Thermal paste to Solder it would make a minimal difference to the experience of the majority of Intel users.

Soldering CPU dies to an IHS is a complex process and CPU dies are exceptionally fragile. Any increase in yield is great for Intel when producing CPUs in large quantities and in many ways soldering can be considered as a waste to all but enthusiast overclockers.

My question is not how many retail CPUs die as a result of soldering, but how many die at an earlier stage. How many don't make it to retail? Then there is also pricing to consider.

With the question of "why doesn't Intel just use better TIM?" my question is how well does Thermal grizzly and other high-performance TIMs work after several years from the original application? Intel's TIM needs to stand the test of time. I am not sure how well enthusiast TIMs work in this regard, as most users of it replace their CPUs or coolers fairly regularly.

 

[AlienALX]

The simple fact here guys is that the majority of Intel's CPUs are not intended to be overclocked, so even if Intel changed from Thermal paste to Solder it would make a minimal difference to the experience of the majority of Intel users.

Soldering CPU dies to an IHS is a complex process and CPU dies are exceptionally fragile. Any increase in yield is great for Intel when producing CPUs in large quantities and in many ways soldering can be considered as a waste to all but enthusiast overclockers.

My question is not how many retail CPUs die as a result of soldering, but how many die at an earlier stage. How many don't make it to retail? Then there is also pricing to consider.

With the question of "why doesn't Intel just use better TIM?" my question is how well does Thermal grizzly and other high-performance TIMs work after several years from the original application? Intel's TIM needs to stand the test of time. I am not sure how well enthusiast TIMs work in this regard, as most users of it replace their CPUs or coolers fairly regularly.

If I am wrong and it is not done to deliberately hobble overclocking then it is done to save money. Having seen Intel CPUs that bent because they had skimped on whatever they are made of nothing would surprise me Mark, tbh.

Intel once had the luxury of charging whatever the hell they wanted to for a CPU. Now they can not, and that means they are going to skimp on materials as much as they can.

It goes back to corporations again. Soulless entities whose only care in the world is profit, and how to extract every last penny. Blood from a stone and all that.

Just like how companies have pretty much shrunk everything we eat and drink as much as they can. It's all one of the same.