>was 1.5ms delay
>now it's 0.15ms delay
WOW! A whole damn millisecond faster!
>>56278064
Adrian Cook
Other urls found in this thread:
Easton Reed
>... 10x faster
Daniel Scott
The point is you wouldn't notice even if it were 100000x faster.
Eli Davis
If all you do is browse facebook (which I assume you do based on your reaction) of course you wont notice it.
Aaron Parker
SSD mad memer detected
Luis Morgan
>1.9 million IOPS
Can't wait for this thing to hit the market.
Jayden Reed
I dare you to provide me an example of activity where 1ms decrease in latency matters.
Cooper Ramirez
>More expensive
>Not significantly faster
>Requires PCIe drivers which are finicky at best
Yes, this truly is the end of NAND. Now watch as no one mentions it again until Intel release it in DIMM, at which point the media will realise that it requires a special Xeon processor to work and it will be quickly recognised as the second coming of RAMBUS.
Hunter Hall
Complains about 1.5 ms delay plays online games with 50 ms delay...
Daniel Murphy
Storage is slow memory.
If you don't understand this then don't bother replying.
Logan Williams
stay mad poorshit
Adam Long
AYYYMDPOORS ON SUICIDE WATCH
Charles Barnes
its 1.5s vs .15s or 1500ms vs 150ms you dumb twat and it means a whole fucking lot for a database or bigdata computation application.
Landon Anderson
How much is this going to cost and will I need a new motherboard to use it as a boot drive?
Dominic Reyes
How can you use anything else if this xpoint shit isn't even out yet.
Zachary Allen
>1.5s
Clear your eyes of semen, retard, it's "us" - MICROseconds. Not milli, not seconds, it's fucking microseconds.
Jack Moore
kys
David Lee
oh, pardon me, you're right.
the result is the same though, it's ten fuckin times faster.
Chase Russell
>DA TEN TIMES FASTA!
Grayson Peterson
Everyone please kys your fucking self
>Yourself your fucking self!!
Grayson Williams
what exactly does this mean?
suicide yourself
Brayden Edwards
Not much. This memory, which is just Micron's phase change memory under intel branding, has lower access latency than NAND.
It can make for a technically faster SSD, but in terms of real world performance its such a insignificant factor to system performance in virtually all metrics that it doesn't matter.
I can't think of many datacenter workloads where this would be significant either. Would it provide some uplift in performance? Sure, but not enough to justify throwing out a million bucks in NAND drives.
Ryan King
>microseconds
diminishing returns
Leo Torres
Climate data simulation.
Heliophysics data processing. Magnetospheric model processing.
The list goes on. See ccmc.gsfc.nasa.gov