Date: Wed, 13 Nov 1996 00:13:50 -0800 From: Chris Browning <cbrown@aracnet.com> To: "Michael L. VanLoon -- HeadCandy.com" <michaelv@MindBender.serv.net> Cc: "Rodney W. Grimes" <rgrimes@GndRsh.aac.dev.com>, scrappy@ki.net, smp@csn.net, smp@freebsd.org, current@freebsd.org Subject: Re: GigaByte GA-586DX-512 Motherboard Message-ID: <3289833E.571F@aracnet.com> References: <199611130647.WAA08610@MindBender.serv.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Michael L. VanLoon -- HeadCandy.com wrote: > > >> >>I recommend 64meg of parity. Don't really have direct knowledge of EDO vs not. > > >> > Hrmmm...I had thought that EDO RAM was better because it had > >> >'on chip' cache or something like that ... > > >> Uh, no. EDO RAM simply takes less clock cycles per access than > >> standard (or FPM) RAM. > > >You should go read some technical specifications before you make such > >statements. The actual access time changes very very little between > >EDO (``Extended Data Out'') and FPM (``Fast Page Mode''). The only > >real effective difference is that EDO memory holds it's data outputs > [...] > > No thank you. I have enough to keep me very busy already. > > I am not a hardware engineer. I understand the _effect_ of running > EDO RAM, and that is that cycle times can generally be shorter, in > typical modern motherboards. > > You can argue about the electical intricacies if you want, but I have > better things to do. I'm sure you know what you're talking about, and > I may get around to looking at this info some time in the future (I am > always curious about the intricacies of the hardware, and generally > understand it fairly well). But right now, I don't have the time. > > How am I wrong in stating that "the effect of running EDO RAM in a > typical modern motherboard is that you can run with shorter cycle > times"? Ok, I will try to answer this. I recommend that we take any future replies off the list. What the previous gentleman was telling you is correct. Both of your statements are incorrect. The cycle time and # of clock cycles to access EDO and FP RAM is approximately the same. Maybe there is confusion here because you are using some very specific terms used to spec memory but I don't think that is what you are intending to say. The cycle time of memory is how fast you can access the same cell of memory sequentially. This is pretty much the same for EDO & FP. The number of clocks that it takes to access memory will be about the same for EDO & FP. Both of these items are determined by the technology used to implement the chips on the S(D)IMMS, which will be the same for both EDO & FP. Both of these really don't talk about the differences between EDO and FP. What makes EDO RAM attractive is the fact that once you have paid the penality to get to memory, you can get a whole row without much additional overhead. This is a kind of "bursting", if you will. This way, if you are getting a lot of sequential bytes, then you pay one large delay to get to memory and a small subsequent delay to get to the adjacent bytes. With FP, on the other hand, you would incur the same large delay to get all the bytes (assuming no interleaving). Unfortunately, some of the advantage of EDO is hidden by the cache. If you have to go to memory, the processor will get the byte (word, etc) that it wanted and the EDO fill will all go to the cache. This makes for faster cache fills, but the processor didn't really see any of the advantage. Well, I hope this help. Again, lets take this off the list. Thanks, Chris
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3289833E.571F>