From the article -<p>> They claimed 10x the density of DRAM, it is now 4x<p>> Latency missed by 100x, yes one hundred times, on their claim of 1000x faster, 10x is now promised<p>> More troubling is endurance, probably the main selling point of this technology over NAND. Again the claim was a 1000x improvement, Intel delivered 1/333rd of that with 3x the endurance.<p>From this seminar few months back - <a href="https://www.youtube.com/watch?v=hXurTRtmfWc" rel="nofollow">https://www.youtube.com/watch?v=hXurTRtmfWc</a> ,<p>I think density can be increased, this is only the initial product,<p>and latency is contributed more by PCIe/OS/application rather than the underlying 3d-xpoint material. The slides from the article are for the PCIe SSDs, I wonder whether the earlier claimed latency, still holds well with NVRAM.<p>I wonder why the endurance is so lower than the earlier claims.
One should not confuse 3D XPoint that is a sort a memory and Optane that is a product, a SDD. Octane uses all the traditional protocols to work like other devices in a computer and that slow down the operations a lot. We saw the same drawbacks with SDD when they used the SATA interface and the protocols of hard drives, before special drivers improved their speed.
How is it "broken"? I thought they meant that there's a fundamental flaw with the tech, but instead it seems he just means that they didn't meet their original goals of the tech with their first actual product.
The new numbers Intel is claiming, can be met by floating gate transistors that are randomly accessible (versus those arranged in bit lines and word lines like in NAND). Especially for a single level cell. No need for another materials science/physics like "x point", unless"x point" actually is just a PR spin without any new physics.
Intel will probably end of spinning this as the first (more impressive) set of numbers being estimates of the potential limits of the new technology.<p>Almost certainly those nice round numbers came from some engineers (or more likely, engineering manager) being pressured for 'long term performance estimates of the technology', coming up with something that seemed plausible, and then wrapped up nicely into a marketing presentation.<p>Or, it might be that intel actually did have something approaching each of those three claims - in three separate embodiments of the product, and is running into problems combining the traits together.<p>In all cases, I feel for the Intel engineers working on this project right now. Probably all cursing that original reveal.