Muranyi & Harjeet, Thanks for correcting me and sorry for causing confusion. I had misinterpreted the PIN R,L,C with die parasitics. Thanks for pointing that out. VS --- "Muranyi, Arpad" <arpad.muranyi@xxxxxxxxx> wrote: > Sorry for the unreadable post, I sent it UTF-8 > encoded > in an attempt to avoid those funny characters, but > it > didn't seem to succeed. Hopefully this one > (US-ASCII) > will. > > Arpad > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D > -----Original Message----- > From: Muranyi, Arpad=20 > Sent: Monday, September 20, 2004 11:29 AM > To: si-list@xxxxxxxxxxxxx > Subject: RE: [SI-LIST] Re: IBIS package R,L,C values > > I would like to correct a comment below which said > that the RLC > values on the [Pin] list will not have priority over > the RLC > values listed in [Package]. The best way to do this > correction > is to quote the IBIS specification, which says: > > |=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > =3D=3D=3D=3D=3D=3D > | Keyword: [Package] > | Required: Yes > | Description: Defines a range of values for the > default packaging > | resistance, inductance, and > capacitance of the component > pins. > | Sub-Params: R_pkg, L_pkg, C_pkg > | Usage Rules: The typical (typ) column must be > specified. If data for > the > | other columns are not available, > they must be noted with > "NA". > | Other Notes: If RLC parameters are available for > individual pins, > they can > | be listed in columns 4-6 under > keyword [Pin]. The > values > | listed in the [Pin] description > section override the > default > | values defined here. Use the > [Package Model] keyword > for more > | complex package descriptions. If > defined, the [Package > Model] > | data overrides the values in the > [Package] keyword. > | Regardless, the data listed under > the [Package] keyword > must > | still contain valid data. > |----------------------------------------------------------------------- > ------=20 > > From this it should be clear that the more detailed > data will > override the less detailed data if available. This > includes the > RLC in the [Pin] list sections as well as the > [Package Model]. > > The purpose of the [Package] keyword was to provide > an overall > description of the entire package. This is why it > has typ., min., > and max. entries. These values can represent the > shortest, longest > and average length lead in the package. Note that > the [Pin] list > section doesn't have typ., min., max. values, it > only allows one > values per pin. This is because the thinking at the > time this > was written in the IBIS spec was that the > manufacturing variations > of the lead frame are so small that they are > negligible, and the > meaning of each individual pin's RLC was to refine > the overall > RLC provided in the [Package] keyword. > > The [Package Model] keyword was designed to make a > more detailed > lumped and/or distributed description of the package > available. > Obviously, this is even better than the lumped RLC > on the [Pin] > list, so if available it should be used instead. > > However, even this syntax has serious limitations > for today's > high speed simulations, so we developed a new and > much better, > general purpose interconnect specification, the ICM > 1.0. This > can be used for connectors, PCB traces, packages, > i.e. any > passive interconnect. I hope some day we will even > use it for > modeling the on die passive interconnects with this > ICM. > > Thanks, > > Arpad Muranyi > Intel Corporation > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > -----Original Message----- > From: si-list-bounce@xxxxxxxxxxxxx > [mailto:si-list-bounce@xxxxxxxxxxxxx] > On Behalf Of V S > Sent: Friday, September 17, 2004 8:17 AM > To: si-list@xxxxxxxxxxxxx > Subject: [SI-LIST] Re: IBIS package R,L,C values > > Mani, > > Every chip has two parts the package and silicon. > The > silicon occupies a small area, let us say, 200 mils > x > 200 mils. While the complete package may be 1 inch x > 1 > inch. There is an interconnect between the silicon > and > the package. This package is modeled using package > R,L,C values. The package interconnect can also be > modeled as a distributed element using ebd files. > The > silicon itself have parasitics, mostly capacitive > load. Both the Pin R,L,C and the package R,L,C > values > are should be taken into account (Most simulators do > it) when simulating.=20 > > > In your case the package will not be taken into > account in case 2. You should see a box outlining > the > driver receive when modeling as "ibis device". > > The difference in the two case is usually small - > especially if the mismatch between the package > impedance and the board impedance is not > significant. > But it will also depend upon the rise time. The > package impedance must be taken into account if you > are working on simulation correlation. Usually the > measurement are done on the motherboard that is > subtantially far from the actual die ( may be as > much > as 1"). In such scenario, the signal at the die is > substantially different from the signal at the die > receiver. In such cases, it may be necessary to take > into account teh package R, L, C values. > > =20 > Some Comments on Harjeet's replies > > The PIN R, L, C values have priority over the > package R,L,C values. > > - I do not think there is anything like Pin PIN R, > L, > C values having priority over the package R,L,C > values. Both are taken into account.=20 > > PS: These R, L, C value are lumped values. For fast > edge rate these may not provide accurate results. > > - It it true that these values are lumped valued. > But > they are accurate in most cases. The vendors may > provide the ebd files that have distributed R,L,C > values rather than the lumped values. I beleive > that, > if a vendor thinks that lumped R,L,C values are > "sufficiently accurate" for the frequency of concern > , > it is ok. > > > VS > > === message truncated === __________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - 100MB free storage! http://promotions.yahoo.com/new_mail ------------------------------------------------------------------ To unsubscribe from si-list: si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field or to administer your membership from a web page, go to: //www.freelists.org/webpage/si-list For help: si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field List FAQ wiki page is located at: http://si-list.org/wiki/wiki.pl?Si-List_FAQ List technical documents are available at: http://www.si-list.org List archives are viewable at: //www.freelists.org/archives/si-list or at our remote archives: http://groups.yahoo.com/group/si-list/messages Old (prior to June 6, 2001) list archives are viewable at: http://www.qsl.net/wb6tpu