Intel 18c HEDT Die Pictured, Not Soldered (7980XE)

[MageTank]

2 minutes ago, Senzelian said:

Don't worry, I validate with Tetris.

 

So, I should spend time to overclock my CPU and in turn you would believe me?
Doesn't seem like a good trade and doesn't convince me.

If it was previously stable at 4.9 on a weak air cooler, surely it would be stable on a custom loop. Simply dial in the previous voltage and have at it. Technically speaking, prime95 won't lag your games or other applications if you intend on multitasking while stressing. Not that u recommend prime95, you will likely cook that poor thing under 48k fft, lol. 

[Senzelian]

9 minutes ago, MageTank said:

If it was previously stable at 4.9 on a weak air cooler, surely it would be stable on a custom loop. Simply dial in the previous voltage and have at it. Technically speaking, prime95 won't lag your games or other applications if you intend on multitasking while stressing. Not that u recommend prime95, you will likely cook that poor thing under 48k fft, lol. 

Hah, I wish I still knew which voltage I was using.  I bought the 6700K in April.2016

Maybe I'll try to overclock it tomorrow - depends on my mood.

[MageTank]

2 minutes ago, Senzelian said:

Hah, I wish I still knew which voltage I was using.  I bought the 6700K in April.2016

Maybe I'll try to overclock it tomorrow - depends on my mood.

If you want to save time, it only takes 5 minutes of Linpack MKL to impress me

 

Spoiler

Nobody should ever do this, not unless you are extremely certain you can handle the worst heat any CPU will ever experience. Delidded on water, I still get butterflies in my gut every time I run it.

 

[dexT]

1 hour ago, MageTank said:

Not unless you validate stability with minesweeper, lol. 

Hey that's what I use err.. I mean Pifast and wPrime

[Jito463]

1 hour ago, MageTank said:

Not unless you validate stability with minesweeper, lol. 

Minesweeper is the best stability tool.  All those mines exploding produce a lot of heat.  

[Senzelian]

4 minutes ago, MageTank said:

If you want to save time, it only takes 5 minutes of Linpack MKL to impress me

 

  Reveal hidden contents

Nobody should ever do this, not unless you are extremely certain you can handle the worst heat any CPU will ever experience. Delidded on water, I still get butterflies in my gut every time I run it.

 

I can walk for 5 minutes in a straight line without falling over after I had half a beer.
That is impressive.

[done12many2]

One of my favorite things on LTT are the plethora of bold OC claims supported by a CPUz screenshot.  lol

[PCGuy_5960]

1 minute ago, done12many2 said:

One of my favorite things on LTT are the plethora of bold OC claims supported by a CPUz screenshot.  lol

iz mi overlock gud?

[done12many2]

2 minutes ago, PCGuy_5960 said:

iz mi overlock gud?

 

Yes as you've demonstrated a lot more than others, so your are definitely good.  Seems completely legit.  Congrats!

[VagabondWraith]

1 hour ago, MageTank said:

If you want to save time, it only takes 5 minutes of Linpack MKL to impress me

 

  Reveal hidden contents

Nobody should ever do this, not unless you are extremely certain you can handle the worst heat any CPU will ever experience. Delidded on water, I still get butterflies in my gut every time I run it.

 

Linpack MKL - 7980XE @ 5GHz.  

[Marinatall_Ironside]

On 9/5/2017 at 1:22 PM, Stefan1024 said:

It has 2 PCBs, that's interesting. Did the older generations had this too?

The two-PCB thing started back with Haswell-E. The last generation of HEDT CPUs to have only one PCB was Ivy Bridge-E.

Thanks to @PCGuy_5960, the two PCB thing started with Broadwell-E. Haswell-E has only one PCB, as there's Google images of Haswell-E delidded 

Edited September 9, 2017 by JurunceNK
Invalidated original statement; Haswell-E has only one PCB.

[PCGuy_5960]

13 minutes ago, JurunceNK said:

The two-PCB thing started back with Haswell-E. The last generation of HEDT CPUs to have only one PCB was Ivy Bridge-E.

Nope, Haswell-E has 1 PCB, Broadwell-E has 2

 

[Marinatall_Ironside]

1 minute ago, PCGuy_5960 said:

Nope, Haswell-E has 1 PCB, Broadwell-E has 2

 

Ah you're correct. Will edit my original post.

[Drak3]

1 hour ago, done12many2 said:

One of my favorite things on LTT are the plethora of bold OC claims supported by a CPUz screenshot.  lol

I just got a Rift, otherwise I'd claim 10GHz on my 5930K and upload a poorly editted CPU-Z screenshot.

[DuckDodgers]

 

Looks like even the top server Xeons for the new LGA3647 socket now come with the cheap paste.

[done12many2]

12 minutes ago, DuckDodgers said:

 

Looks like even the top server Xeons for the new LGA3647 socket now come with the cheap paste.

 

Thanks for sharing that.  Should help some of the engineers here.  

[mynameisjuan]

16 minutes ago, DuckDodgers said:

 

Looks like even the top server Xeons for the new LGA3647 socket now come with the cheap paste.

Oh lawdy now all the server admins are going to riot with their pitch forks about this BS!!!!...../s

 

Like come on people. TIM only matters for OCing which for one is not done on Xeons by almost everyone. You are just trying to make this a problem.

[Drak3]

3 minutes ago, done12many2 said:

 

Thanks for sharing that.  Should help some of the engineers here.  

Speaking as an engineer of destruction, it does.

[PCGuy_5960]

30 minutes ago, DuckDodgers said:

 

Looks like even the top server Xeons for the new LGA3647 socket now come with the cheap paste.

This basically proves that Intel couldn't solder these dies. Xeons were always soldered

[Drak3]

2 minutes ago, PCGuy_5960 said:

This basically proves that Intel couldn't solder these dies. Xeons were always soldered

No, it proves that Intel has reason to use TIM over solder, likely a very good reason.

[mynameisjuan]

7 minutes ago, Drak3 said:

No, it proves that Intel has reason to use TIM over solder, likely a very good reason.

Same reason they do on desktops. A majority of chips sold are not over clocked and dont need to work about extreme temps. Especially with xeon almost never being OCed.

[Drak3]

Just now, mynameisjuan said:

Same reason they do on desktops. A majority of chips sold are not over clocked and dont need to work about extreme temps. Especially with xeon almost never being OCed.

AVX is used significantly more in the Enterprise. Xeons are seeing higher temps than Core i counterparts being ran at stock. There is benefit to soldering Xeons that will run at stock

[Jito463]

3 hours ago, mynameisjuan said:

Like come on people. TIM only matters for OCing which for one is not done on Xeons by almost everyone. You are just trying to make this a problem.

Soldering helps with temps, and temps are important for any computer; but even more so with servers where typically multiple chips are run on multiple servers simultaneously.

3 hours ago, Drak3 said:

AVX is used significantly more in the Enterprise. Xeons are seeing higher temps than Core i counterparts being ran at stock. There is benefit to soldering Xeons that will run at stock

Exactly, temps aren't just important because of over clocking.

 

Now, maybe there's a legit reason for Intel to not solder these chips, or maybe it's nothing more than a cost cutting measure.  Frankly, I'm more inclined to believe the latter, but I'll acknowledge it could be the former.

[mynameisjuan]

3 minutes ago, Jito463 said:

Soldering helps with temps, and temps are important for any computer; but even more so with servers where typically multiple chips are run on multiple servers simultaneously.

yeah but at stock what is the temp difference? I know temp is important and also for sound but server fan run balls to the wall anyway no matter temps.

[Jito463]

2 minutes ago, mynameisjuan said:

yeah but at stock what is the temp difference?

The exact numbers?  I couldn't say.  I don't run servers personally, and I don't have access to the necessary hardware to test.  However, even a few degrees difference can help significantly with their cooling solutions, and could save hundreds - if not thousands - of dollars per month in power for said cooling.