They don't talk about specific numbers because apparently, the nature of the pipeline has the latency vary over time under load, so HBM may be initially slower under light load it remains more consistent under heavy load so, in the end, it is faster for heaver applications.I started listening to the first few minutes of that video and realized it wasn't getting into any specifics and then started quickly skipping through it.
Did they mention any numbers or did they specifically say that HBM memory has much faster or slower latency vs GDDR or DDR? If so can you give me the timestamp. Thanks.