From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from server.bluewater.co.nz ([203.167.251.250] helo=smtp.bluewaternz.com) by pentafluge.infradead.org with esmtp (Exim 4.14 #3 (Red Hat Linux)) id 19QZyb-0005pU-NW for ; Thu, 12 Jun 2003 22:51:22 +0100 Received: from benmore.bluewaternz.com ([192.168.2.3] helo=bluewatersys.com ident=gordon) by smtp.bluewaternz.com with esmtp (Exim 3.03 #1 (Debian)) id 19QZz4-0003Gy-00 for ; Fri, 13 Jun 2003 09:51:50 +1200 Message-ID: <3EE8F5F6.7050805@bluewatersys.com> Date: Fri, 13 Jun 2003 09:51:50 +1200 From: Gordon J Milne MIME-Version: 1.0 To: linux-mtd@lists.infradead.org References: <20030612163736.CFEC.MINGFENG@linpus.com> In-Reply-To: <20030612163736.CFEC.MINGFENG@linpus.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: MRAM will kill FLASH ? List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , mingfeng wrote: > who know MRAM(Magnetoresistive Random Access Memory) ? > Will it kill FLASH soon? Hopefully it will kill FLASH but I cannot see it happening anytime soon. MRAM storage densities have a long way to go to make MRAM a competitive alternative to FLASH. But this is to be expected because the MRAM developers have only just moved form the working on the device fundamentals to the production building issues. It is difficult to say how long it will take but I suspect that as soon as device densities approach 50% that of FLASH, you will see MRAM taking over in specialist devices - top end PDAs, niche embedded systems, etc. And, of course, the Instant-On computer. The recent announcement by Toshiba and SanDisk about futher reducing the size of NAND FLASH cells increases the lifetime of FLASH and raise the bar for MRAM deployment. BTW, a recent issue of IEEE Spectrum had a good overview article on the "new" memory technologies of MRAM, FRAM (Ferromagnetic RAM) and OUM (Ovonic Unified Memory). It described the benefits of each and the underlying physical principle involved. See http://www.spectrum.ieee.org/spectrum/mar03/features/semi.html fro more details.