Discussion Packaging: EMIB, Infinity Fabric, CoWoS, Foveros

[DrMrLordX]

I think that in general, Intel may be forgetting what got them here. Truly having an open ecosystem where others could add value to that ecosystem, and that the platform is a key part of the success of the x86 market. Intel still talks about it in that way, but that's not what they are doing, any pico-acre of silicon that doesn't belong to Intel is something that they covet. But I think acting that way is to the detriment of the health of their platform ecosystem long term.

Intel jumped ship on open ecosystems in 1997. I still loathe them for it.

 

[IntelUser2000]

3DXPoint is NOTHING EVEN REMOTELY CLOSE like what was announced and hyped for 2 years.

Yes the SSD version, but with the DIMM they are pretty much as promised on the performance front. Endurance isn't but its good enough to write 24/7 at max write speeds without it reaching the limit(because its 15-30x better than Optane SSDs).

Plus, they'll be fully pin-compatible in the DDR5 generation(3rd gen Optane). Also there's this:
https://www.pcworld.com/article/310...mory-to-work-on-amd-pcs.html#tk.rss_computers

Intel wants to make adoption of Optane easy for makers of PCs and servers regardless of the chips they use, said Rob Crooke, senior vice president and general manager of the company's Non-Volatile Memory Solutions Group, in an interview.

So let's hope they live up to that promise.

 

[lobz]

That's a pretty strong exaggeration. You could say 3DXpoint products aren't what was promised, but the technology itself appears to be exactly what was promised. And I expect that at some point you'll see products getting quite a bit closer to those theoretical performance numbers.

I understand your point and normally I'd agree, but the misleading way it's been presented was much stronger than my exaggeration. In my humble opinion of course

 

[moinmoin]

Also there's this:
https://www.pcworld.com/article/310...mory-to-work-on-amd-pcs.html#tk.rss_computers

So let's hope they live up to that promise.

Sounds promising, until I looked at the date the article was published ("AUGUST 09, 2016"). Since then they made more of an exclusivity push since Intel as a whole has not enough to show due to their 10nm troubles.

 

[Det0x]

Nevermind, already posted

 

[Gideon]

but with the DIMM they are pretty much as promised on the performance front

According to AMD's Forrest Norrod it isn't really. the best case latency is ok-ish but it's non-uniform so I'd really like to see the worst case. DDR latency is uniform, so unexpected things can happen with Optane.

So the other aspect of course is lower cost per bit. They’ll use it as DRAM replacement and the fact is that it has longer latency and non-uniform latency, and so there are a bunch of issues there.

 

[IntelUser2000]

According to AMD's Forrest Norrod it isn't really. the best case latency is ok-ish but it's non-uniform so I'd really like to see the worst case.

I'm referring to the post about Optane not meeting original claims. Read the previous posts to get some context.

Also, they never said it'd be fast as DRAM. But the DIMM is pretty close.

3rd generation will be pin compatible with DDR5, so you might never know if all it needs is the CPU to have a compatible controller(like AMD server CPU of that timeframe).

 

[Gideon]

I'm referring to the post about Optane not meeting original claims. Read the previous posts to get some context.

Fair enough

3rd generation will be pin compatible with DDR5, so you might never know if all it needs is the CPU to have a compatible controller(like AMD server CPU of that timeframe).

IMO It's almost assured, that Micron will provide it's version of Optane to AMD anyway, so Intel might follow suit