Thursday, August 10th 2017

AMD Ryzen Threadripper "Summit Ridge" Dies are Heavily Binned

AMD revealed that the pair of 8-core "Summit Ridge" dies that make up the Ryzen Threadripper multi-chip module are heavily binned. AMD hand-selects the top-5% highest performing "Summit Ridge" dies for Ryzen Threadripper manufacturing, which makes these chips of a higher grade than even what AMD sets aside for Ryzen 7-series socket AM4 chips.

AMD requires the highest grade "Summit Ridge" dies to use in Threadripper chips, to keep electrical leakage to the minimum, so the chips can run as cool as possible, with the least power-draw. Choosing the best dies could also ensure that Threadripper chips have the highest overclocking-headroom taking into account other electrical and thermal constraints. A 7-series chip such as the 1800X could still achieve higher clocks than a Threadripper chip, in that sense.
Add your own comment

60 Comments on AMD Ryzen Threadripper "Summit Ridge" Dies are Heavily Binned

#26
fritoking
qubitHow do you know they're blank silicon and not disabled cores?
I don't have the sources, but I read an article a week or so ago and Amd stated they were blank ...no possible unlocking due to that fact.
Posted on Reply
#27
dwade


And then Arctic Coolers think their tiny heatsinks will be able to cool these.
Posted on Reply
#28
jabbadap
cdawallUnless you can microsolder all the missing transistors in place you weren't unlocking anything with his cpu.
Well if that was indeed ES, it could have been just epyc processor and could have been worked on epyc motherboard as 32 core processor without any modifications needed. Of course this are all just hypothetical mumbo jumbo now anyway, retail TRs should be with those blank silicone slaps for equalize pressure for processor cores.
Posted on Reply
#29
cdawall
where the hell are my stars
jabbadapWell if that was indeed ES, it could have been just epyc processor and could have been worked on epyc motherboard as 32 core processor without any modifications needed. Of course this are all just hypothetical mumbo jumbo now anyway, retail TRs should be with those blank silicone slaps for equalize pressure for processor cores.
No it was missing half of the transistors to run the chip around the edge. That chip wasn't doing anything.
Posted on Reply
#31
jabbadap
cdawallNo it was missing half of the transistors to run the chip around the edge. That chip wasn't doing anything.
You mean after he delid it, which surely destroyed whole thing or? I was meaning before the delidding. It was working processor before delidding to my knowledge.
Posted on Reply
#32
cdawall
where the hell are my stars
jabbadapYou mean after he delid it, which surely destroyed whole thing or? I was meaning before the delidding. It was working processor before delidding to my knowledge.
No I mean that the CPU from AMD only had half of the transistors on it total. It only had power delivery for the two dies that were "functional" there would be no way to make the other dies work regardless of what they were.
Posted on Reply
#33
notb
qubitYes, I can be serious and I really don't like your tone towards me. Always unpleasant so how about you lighten up a bit eh? No need for unpleasantness regardless of who's technical point is right or wrong.

I think Intel will care about the gaming market, even a niche one as it's all about prestige and image, which helps to bring in the sales.
There really is no prestige in the HEDT segment. Actually it's not very well known outside of communities like this one. We're so far in the notebook revolution, most people won't even be able to name the LGA1151 socket. Or the last time they've seen a desktop CPU live...
Intel makes almost no marketing campaigns about X299 (same with X99 before). Their "flagship" CPU for gaming is 7700K. Huge Xeon lineup covers workstations and servers. HEDT lineup is mostly to suck some money from geeks and recycle some failed Xeon dies.
This is also why previous HEDT CPUs had bonkers pricing. They really didn't care whether the CPUs sell or not.
Intel won't invest big money on bringing out a new platform they "don't care about", that's just silly. More likely they're not sure how to react to the Threadripper threat which is why we have these anomalies with Skylake-X and the next generation will be much better.
But they haven't invested big money into X299. This is pretty obvious once you look at the specs... :)
They just used things they had. Skylake and KabyLake are so good they actually might release one more generation just by tweaking and clocking higher - we know this because there is still some decent OC headroom (unlike Ryzen which is clocked near the limit).
New high-end Xeons, on the other hand, are pretty advanced. It's a first "socket" implementation of Knights Landing - something we could actually call "next-gen" CPU.
How can you say for certain that AMD won't allow their CPUs to be unlocked? On what authority are you basing that on, or is that just your opinion? I'm saying maybe/hopefully and that's the best we can do at the moment.
Because the other 2 modules are not functional dies.
cdawallYes it is 8 channels.
Well... it's 8 channels sort of... It's 2 channels per die.
Posted on Reply
#34
thesmokingman
fritokingI don't have the sources, but I read an article a week or so ago and Amd stated they were blank ...no possible unlocking due to that fact.
Here:

www.extremetech.com/computing/253416-amd-explains-threadripper-cpus-four-die-hood
The explanation is straightforward: The reason Threadripper has four dies under the hood with only two in use is because leaving them off would create mechanical instability when the heatsink is mounted. AMD apparently needs four dies in that arrangement to stabilize the packaging. This does make sense, given some of the esoteric cooling solutions with high pressure attachments that overclockers use...

The other important thing to know about Threadripper is that the other two die apparently aren’t actually chips at all. They’re apparently structural inserts required for support, but do not represent “bad” Epyc cores or any kind of core recycling. OC3D also states that the two active die in Threadripper are on the diagonal from each other.
Posted on Reply
#35
cdawall
where the hell are my stars
notbWell... it's 8 channels sort of... It's 2 channels per die.
They are working like 8 channels in bandwidth just fine from testing that is already available. This is a stark difference from what I expected and how the older AMD server chips worked
Posted on Reply
#36
jabbadap
cdawallNo I mean that the CPU from AMD only had half of the transistors on it total. It only had power delivery for the two dies that were "functional" there would be no way to make the other dies work regardless of what they were.
Well x399 platform does not have power delivery for it, but what about sp3 platform. My point just was: if that was just epyc processor marked as Threadripper, it might have the all needed circuits untouched on die and those two cores were just powered down when not in use by x399 platform. Of course if someone is stupid enough it could slap epyc processor to TR4 socket(physically it should fit) and look what happens. It most probably wont boot and might potentially destroy them both(mobo and cpu), but more likely because of missing microcode for processor.
Posted on Reply
#37
erocker
*
XiGMAKiDAMD and chips in the same sentence remind me of this
Lol, looks like they took a Pringles can, made it look almost like a Campbell's Soup can and stamped AMD on it. AMD is weird.
Posted on Reply
#38
trparky
Ah yes... both Threadripper and Epyc get the very best silicon whereas us mere peasants buying the Ryzen chips get the... less than desirable stuff. The peasants get the table scraps, the fat that was trimmed off, etc.
Posted on Reply
#39
infrared
trparkyAh yes... both Threadripper and Epyc get the very best silicon whereas us mere peasants buying the Ryzen chips get the... less than desirable stuff. The peasants get the table scraps, the fat that was trimmed off, etc.
Same with high end intel chips, they're just the dud chips that would have otherwise been a Xeon. A bit irritating from our point of view but makes sense for the businesses. The TR/Epyc/Xeon cpus need to have as little electrical leakage as possible to keep tdp and temps down. On the mainstream cpu's this isn't as critical, us overclockers are just extremely fussy about our sillycon!
Posted on Reply
#40
trparky
Oh I know this, I do indeed know this. Heck, some chips even in the mainstream lineup as better than other chips hence the "Silicon Lottery" that a lot of folks talk about.

My CPU for instance, I won the "Silicon Lottery" when I bought my chip.
Posted on Reply
#42
trparky
Wow, just wow @PowerPC . That's a whole lot of awesome right there!
Posted on Reply
#44
xorbe
What memory speeds are achievable with 4x16GB? I think all the reviews used 4x8GB today.
Posted on Reply
#45
Hood
Durvelle27Now I want to upgrade
Don't feel bad just because the top 5% were used for TR/EPYC, that still leaves 95% for Ryzen's silicon lottery. Your 1700 could still be in the top 80-95% (you have to determine that yourself by experimentation). Other people with the same CPU will eventually post their highest stable clocks and lowest voltage required to get there, along with benchmark scores, and you'll be able to see how you did in the lottery. If it turns out that your 1700 sucks compared to others, then it might be time to upgrade.
Posted on Reply
#46
Dave65
WTG AMD!
I wonder if the gaming performance will increase with better drivers, like Ryzen did..
Posted on Reply
#47
Doctor Who
This is great news. This guarantees the perfect CPU when purchasing.
Posted on Reply
#48
Durvelle27
HoodDon't feel bad just because the top 5% were used for TR/EPYC, that still leaves 95% for Ryzen's silicon lottery. Your 1700 could still be in the top 80-95% (you have to determine that yourself by experimentation). Other people with the same CPU will eventually post their highest stable clocks and lowest voltage required to get there, along with benchmark scores, and you'll be able to see how you did in the lottery. If it turns out that your 1700 sucks compared to others, then it might be time to upgrade.
It can do 4GHz

But I want to upgrade for the cores
Posted on Reply
#49
Frick
Fishfaced Nincompoop
dwade

And then Arctic Coolers think their tiny heatsinks will be able to cool these.
Don't overclock and I bet they'll work.
Posted on Reply
#50
Doom2pro
The article incorrectly states that low leakage (high binned) chips are easier to overclock, but if you talk to actual overclockers, they all pretty much agree that leaky chips overclock better because the transistors parasitic components are less effective at undermining the transistors switching speed.

This is why in GPU ASIC Quality, a higher ASIC quality (which has less leakage and lower power per clock) die has less overclock headroom because a higher ASIC quality means higher quality transistor means higher quality transistor parasitic properties which means more difficulty clocking higher.
Posted on Reply
Add your own comment
Nov 21st, 2026 03:29 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts