any solid data on 25, 34 and 50nm NAND?

i'm curious as to why 34nm is better than 25nm and why ssd makers are shying away from 50nm besides cost. SLC drives have their uses, but the cost is astronomical compared to the already expensive MLC ssd's. and yeah we need a sticky phil. hint hint. hurry up with your "tests".

 

34nm has an expected write cycle of about 5000, while 25nm has about 3000, IIRC. Write cycles are what will end the ssd. Disclaimer, although I have been reading about this for over a year, I am still waiting for the feedback flood of ssd's dying because of the write cycle being exceeded. The SLC drives are enterprise, meaning they are able to be used for purposes like servers, and have much higher write cycle life. Again IIRC, up to 100,000 write cycles. That is why they are so expensive.

 

It's a bit faster and last longer (5000 vs. 3000 write cycles).

I shouldn't have said anything in the first place It's still unknown when and where it will be published. Don't expect anything shocking though. Hardwareheaven has a good review online.

I agree on the stickies but that's not up to me.

 

There was a certain website which did some full on write tests. I can't find it now, but if I recall correctly they did constant writes to X25-m and Toshiba HG2, the Toshiba of which went to read only mode after eight months?

edit:
http://www.thosp.com/PC/SSD_vs_HDD/RBER/RBER_en/

 

How old is this? Intel V is first gen, no? The Toshiba is a 43nm, the Intel is 34nm. Both are Indellinx controller run (if I read the link correctly). I have 3 Intels and a Corsair Nova. The Nova is the only one I have with the Indellinx controller. I am impressed by the write cycles. Both went 3 times what everything I read says they are going to get. If that is the case, I will be very, very happy.

 

Intel has never used Indilinx controllers as far as I am aware; all but one of their drives uses in-house Intel controllers (and the one time they used another controller was because they wanted to get a SATA III product out more quickly).

 

Yes the Intel uses an Intel controller, the Toshiba uses a Toshiba controller based on a JMicron controller.

 

Yeah, read the link he provided, if I read it right, it says that they are both Indellinx controlled. I have an Intel 120gb x25-m, and Intel 310 and a 510, and the 510 is the only one not Intel controlled.

I might have read it wrong... I scanned through it, and after a second reading, it also includes Super Talent, OCZ and Samsung. Too much fine print, and too late for me. But If any of the drives get 17k write cycles, I am impressed. I will actually read the whole article tomorrow, and hit all the inks in it, and try to see what the heck it actually says. About whatever drive it is talking about.

OK, It is only at the very end of the article where it really mentions an Intel, and it is an Intel x25-v, not an Intel x25-m. I scanned though the article thinking that all the tests were on these 2 drives (Intel x25-m/v and Toshiba), but in fact they are on about 6 different drives. I am so lost. Solid Data, A Data, OCZ, Intel, Toshiba, Samsung.... I have no idea anymore what the author was talking about. But I will be very happy if I can get anywhere near the 17-18k writes they talked about.