Leak confirmed: AM4 CPUs will have the IHS soldered

thesmokingman

thesmokingman

Supreme [H]ardness
Joined
Nov 22, 2008
Messages
6,617
And thank goodness too because look at all those ic's lined around the IHS making it near impossible to remove the IHS w/o destroying something in the process.

AMD, in contrast to Intel and its mid-range CPUs strategy (e.g. Skylake-S Socket 1151), will try to tempt Enthusiast users thanks to high quality and fast SKUs, sold at competitive prices. Welcome back, Athlon XP times.
Click to expand...

yJIdZ4Qh.jpg


http://www.bitsandchips.it/52-english-news/7784-our-leak-confirmed-am4-cpus-will-have-th
 
I really hope benchmarks pan out and I can plan an AMD system some time soon! Most of the time we don't delis anymore... so hopefully the ihs is soldered nicely.
 
jmilcher said:
I really hope benchmarks pan out and I can plan an AMD system some time soon! Most of the time we don't delis anymore... so hopefully the ihs is soldered nicely.
Click to expand...

The positives are we save some time, a few bucks maybe and most importantly our warranty remains in tact! There might be some fringe lunatics that go bare die but man, that will be serious work to not damage something.
 
so for those who aren't overclockers, can anyone explain in more detail what this means? How does the IHS affect overclocking?
 
SighTurtle said:
so for those who aren't overclockers, can anyone explain in more detail what this means? How does the IHS affect overclocking?
Click to expand...

Often times the thermal paste that Intel uses is lackluster and means higher temperatures overall. When you already have high temperatures, pre overlock, your overclock headroom is reduced.

This CPU will not have to be delidded because the lid is directly soldered.
 
auntjemima said:
Often times the thermal paste that Intel uses is lackluster and means higher temperatures overall. When you already have high temperatures, pre overlock, your overclock headroom is reduced.
Click to expand...

Adding to the above, the crap paste they use can leave as much as 20c-30c lower temps on the table. Recently like this week, two examples of the 7700K were delidded, had their crap paste replaced with a liquid metal paste and they both saw from 26c to 30c lowered temps. Temps become vitally important when overclocking especially when going for high overclocks. With the standard crap paste, you cannot go for high overclocks because you are already 20c-30c closer to the tjmax, ie. the maximum safe temperature of the cpu. But if you can lop off 30c, you are golden, thus that is why enthusiast overclockers delid.
 
That's a Bristol Ridge APU....And its around 50mm2 bigger than what Ryzen is expected to be. And its not solder either as already explained.

am4_0effective4.jpg

am4_0effective5.jpg
 
thesmokingman said:
Adding to the above, the crap paste they use can leave as much as 20c-30c lower temps on the table. Recently like this week, two examples of the 7700K were delidded, had their crap paste replaced with a liquid metal paste and they both saw from 26c to 30c lowered temps. Temps become vitally important when overclocking especially when going for high overclocks. With the standard crap paste, you cannot go for high overclocks because you are already 20c-30c closer to the tjmax, ie. the maximum safe temperature of the cpu. But if you can lop off 30c, you are golden, thus that is why enthusiast overclockers delid.
Click to expand...

And you compare delidded with lidded ;)
 
Shintai said:
That's a Bristol Ridge APU....And its around 50mm2 bigger than what Ryzen is expected to be. And its not solder either as already explained.
Click to expand...
Pretty sure it's soldered, bud. See the outside of the die, and the imprint on the edges of the solder on the IHS? It is solidified. Also, his post in Korean roughly translates to "To the naked eye, didn't break the core, but can get damaged with high heat, doesn't work" <---soldered

Carrizo / BR die: 244mm^2
Expected SR die: Unknown, I expect <230mm^2

Those shots you posted are of the same CPU, but after the solder was sanded and thermal paste applied. The original photos were posted to Facebook by overclocker Namegt and we were speculating whether it was BR or SR on his facebook post, but he didn't actually say. I put on his post that I thought it might be Bristol Ridge which may have caused some confusion.

Did someone measure the chip size based on the photo?
 

Attachments

  • namegt.png
    namegt.png
    292.8 KB · Views: 70
Last edited:
BeepBeep2 said:
Pretty sure it's soldered, bud. See the outside of the die, and the imprint on the edges of the solder on the IHS? It is solidified. Also, his post in Korean roughly translates to "To the naked eye, didn't break the core, but can get damaged with high heat, doesn't work" <---soldered

Carrizo / BR die: 244mm^2
Expected SR die: Unknown, I expect <230mm^2

Those shots you posted are of the same CPU, but after the solder was sanded and thermal paste applied. The original photos were posted to Facebook by overclocker Namegt and we were speculating whether it was BR or SR on his facebook post, but he didn't actually say. I put on his post that I thought it might be Bristol Ridge which may have caused some confusion.

Did someone measure the chip size based on the photo?
Click to expand...

That chip in the picture is Bristol Ridge. And its not solder but a soft metal like TIM from the looks of it. Its from back in november.

2cca816c85fc.jpg
 
Shintai said:
That chip in the picture is Bristol Ridge. And its not solder but a soft metal like TIM from the looks of it. Its from back in november.
Click to expand...
Again, I'm 99.9% sure that is an image of him applying TIM after removal of most of the solder.
I know its from November, I just uploaded a picture of his facebook post, which I commented on with a picture of the BR die overlayed on the chip, saying I thought it was BR. It seemed to fit within the dimensions however nobody took the time to measure it in comparison with the full package size, so we weren't quite sure. He didn't say. ;)

News sites ran with it after he posted it. If you look closely at the picture the OP posted (from the same set), it looks like the top layer of the edge of the die came up with the solder on the IHS, which is why the chip was dead afterwards.
 
Last edited:
The IHS is clearly gold coated so it's should be pretty obvious that it's soldered. Also there's clearly epoxy everywhere too around the chip. They wouldn't do all that just to use crapola toothpaste between the core and IHS.
 
You must log in or register to reply here.
Back
Top