SSD za sistemski disk

Dr_M je 1. jul 2014 ob 16:09 izjavil:

Seveda sm to cilju na paranoike. :>
Po petih letih in 5 razlicnih pogonih sem naredil tak zakljucek, da se nikakor ne sekiram za endurance.

hojnikb je 1. jul 2014 ob 12:47 izjavil:

http://anandtech.com/show/7896/plextor-...

Traditionally cells are symmetrical. The process node refers to the size of one cell, so in the case of 20nm NAND, the cell size would be 20nm x 20nm. However, there is no rule against making an asymmetrical cell and that is in fact what Toshiba/SanDisk did with their first generation 19nm NAND. Instead of being a symmetrical 19nm x 19nm design, the cell size was 19nm x 26nm. Compared to IMFT's symmetrical 20nm design, the actual cell size is quite a bit larger (494 nm2 vs 400 nm2 ), yet in terms of marketing Toshiba's/SanDisk's "19nm" NAND was smaller and more advanced.

Se bo treba naučit _natačno_ brat

Oftentimes when cell size is discussed, it is only the actual size of the cell that is taken into account, which leaves the distance between cells out of the conclusion. However, the so called empty parts (they are not really empty as you saw in the X-ray a couple of pages back) take a part of the die area similar to the cells, meaning that they cannot be excluded. The appropriate way to measure cell size is from the inner corner of a cell to the outer corners of neighbouring cells in both X and Y axes. This is demonstrated by the black square on the above graph.

Tako da je čisto odvisno kako meriš. Če meriš samo celico, potem node size predstavlja velikost stranice celie. Če pa upoštevaš še ta "prazn" plac med celicami, je pa jasno velikost večja.

Dr_M je 1. jul 2014 ob 15:27 izjavil:

Jest sumim, da pri temu "nalozenemu" nandu uporabljajo vecje kapacitete, kot uradno so, da imajo tako velik endurance. Ok, neki dobijo na racun vecjega proizvodnjega procesa, ampak mocno dvomim, da je to dovolj. Kar tako na lepe oci, samo zato, ker je "3D" pac ne gre.

Dr_M je 1. jul 2014 ob 15:33 izjavil:

Dvomim, da so samo s procesom toliko pridobili. To je 3x vec, kot je bil MLC pred leti podobne proizvodnje, kar nekako ni logicno. Pomoje vec kot 15-20k niso samo s tem dobili.

darkolord je 1. jul 2014 ob 18:59 izjavil:

hojnikb je 1. jul 2014 ob 15:44 izjavil:

SuperVeloce je 1. jul 2014 ob 15:42 izjavil:

od takrat, ko smo imeli 40nm ssd-je pa do sedanjih 1xnm so morali proizvajalci uporabiti tudi polno drugih pogruntavscin, katere se lahko sedaj uporabijo...
po pravici povedano se meni zdi cudez, da se lahko flash programira na tako visokih napetosti s tistimi parimi elektroni na celico in zdrzijo... in da lahko to kupimo za drobiz...

Jap, v TLCju pridejo bojda na celico samo 3je !!! elektroni. Imagine that.

?

Trije biti pa res niso trije elektroni :)

hojnikb je 1. jul 2014 ob 19:35 izjavil:

Trije elektroni na bit-state pri 16 nm, če sem prav razumel. Pač bolano nizke cifre in rata ti jasno, zakaj zadeve lahko crkujejo :)

Jap. Zto je prehod na 3D pratkično nujen, ker skalirat pod 10nm že skor nemorš brez čarovnij.

The scaling introduces a variety of issues but I will start with the cell size. As the cell size is shrunk, the size of the floating gate is also shrunk, which means that the floating gate is able to hold less and less electrons every time the process node gets smaller. To put this into perspective, Toshiba's and SanDisk's 15nm NAND is stores less than 20 electrons per NAND cell. With TLC, that is less than three electrons per voltage state, so there is certainly not much headroom for escaped electrons. In other words, the cell becomes more vulnerable to the IPD and tunnel oxide wear out because even the loss of one electron can be vital to the voltage state.

Ok malo sem se narobe izrazil (ni na celico toliko elektronov, ampak na voltage state). Still, še vedno presneto malo.

hojnikb je 1. jul 2014 ob 20:00 izjavil:

FlyingBee je 1. jul 2014 ob 19:57 izjavil:

hojnikb - Tuesday, July 01, 2014 - link
Your EVO has worse endurance, lower IOPS, lower consistency and most of all lower Write speeds (495MB/s write is due to trick called turbowrite).
Sequential speeds are only a part of the story.

zakaj je že endurance sploh pomemben?

Ne gre se zato, kako pomemben je, gre se zato, da je to ena izmed razlik med tema pogonoma.
Ljudje pogledajo sekvenčne številke (ki so itk napihnjene pri EVOtu) in misljo kako stvar ni nič boljša. Čeprav temu ni tako.

Dr_M je 1. jul 2014 ob 19:56 izjavil:

Ok, sm v sluzbi in na telefonu, tko da ne bom sel iskat, zanima pa me, za koliko je rejtan ta samsungov nand? Omenjeno je bilo 10.000, ampak je res toliko, ali je ta stevilka neznana?

Žal ni nobenga official rejting, vse so samo ugibanja. Glede na nod, ki ga uporabljajo in endurance rejting bi reku da je sam flash uradno dober vsaj za 10k prepisov. Lahko se pa seveda motim.

Between 800 and 900TB of writes, the 840 Series logged 119 more uncorrectable errors, bringing the total to 295. Anvil didn't report any hash failures during that period, but we have its built-in integrity test set to run relatively infrequently--after each 1TB of writes--and on a 700MB file, that covers only a small portion of the flash. Regardless of whether the last spate of uncorrectable errors resulted in incorrect data, it's probably no coincidence the 840 Series died shortly after.

Dr_M je 1. jul 2014 ob 20:13 izjavil:

A to ciljas na 840 pro? Ker evo je poginil ze prej (700 ce se prav spomnim)

Mimogrede, za obicajen nand je ocitno ze 16nm problem, ker ga ni v mx100 128GB. Ubistvu bo vso stvar resevala velika kapaciteta, pri 10nm se bomo pogovarjali ze o vec TB ssdjih.

the endurance rating remains at the same 72TB despite the change in NAND. Micron wouldn't give us the exact P/E cycle rating for their 16nm NAND but I was told that the endurance is similar to their 20nm MLC NAND.

verjetno so poslani s strani proizvajalcev..

Dr_M je 1. jul 2014 ob 20:53 izjavil:

verjetno so poslani s strani proizvajalcev..

Ce je temu res tako, so izsledki analize se vedno verodostojni?

radmannsdorf je 1. jul 2014 ob 14:14 izjavil:

danes bom frendu fliknu notr MX100, in to v komp ki je star 5-6let. notr ma eno hudo plato za tiste čase, pa me zanima, glede na to da seveda ima sata priklope, to tudi avtomatsko pomeni da ima v biosu AHCI možnost ane?

Dr_M je 2. jul 2014 ob 11:09 izjavil:

Pipi, narobe si ti tale 3d predstavljas. Endurance se poveca zaradi drugacnega "zapiranja" elektronov v celico, ne zaradi visanja tranzistorjev.
V tem primeru se gre zgolj za vecjo gostoto, ki gre v visino, endurance je pa zgolj posledica drugacnega shranjevanja elektronov in predvsem 40nm proizvodnjega procesa, ki omogoca zelo velike celice.
Sicer je v clanku na anandu vse lepo razlozeno.

FlyingBee je 2. jul 2014 ob 19:30 izjavil:

http://www.legitreviews.com/samsung-850...

zakaj dokaj nizka hitrost v 4k?