Our company is in the same situation - all our designs are done in the Cadence spectre environment. S2IBIS does support spectre (in your setup file you specify spice or spectre) and I started using this flow but I found a number of non-idealities with it that I started editting and source code (eg to include >100 points per waveform (which is a ver 4.1 feature) which gives me adequate points during the transitions), and editing the simulation files that it would create. After spending several weeks on this, I decided it was easier to just easier to write my own S2IBIS script which would be tuned for Spectre, tuned for 4.1, tuned for our company's types of I/Os (for example some of our I/Os need some initial time to stabilize after power up before measurements commence and S2IBIS didn't provide this feature), and tuned for the features that I want. This took me a month but now I've created many IBIS models with it (that I have simulated in spectre against their "extracted, transistor-level" design to prove their validity) and have had great success. Good luck with whichever method you choose. Dwayne Johnson IC Design Engineer Gennum Corporation, 4281 Harvester Road Burlington, Ontario, Canada L7L 5M4 website: www.gennum.com <http://www.gennum.com/> email: dwayne_j@gennum.com phone: 905-632-2999 x 2640 fax: 905-633-8105 _____ From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Sudarshan Honnudike Sent: Friday, November 10, 2006 6:25 AM To: Muranyi, Arpad Cc: ibis-info@eda.org; ibis-users@eda.org; owner-ibis-users@eda.org Subject: RE: [IBIS-Users] Warnings in V-t curves Hi Arpad, As I am working on only generating IBIS models for our buffers and never looked into SI simulations, it became difficult to me for understanding those parameters. Yesterday I contacted one of our customer and saw how they do SI simulations. That really helped me in clearing some of my doubts that i had regarding Vmeas, Vref, Rref and other parameters. I want to know is there any tool like spectre2IBIS like S2ibis?. I know Spice is universally accepted language and hence only S2ibis is present. But just as a clarification I wanted to know, by any chance is there any tool like spectre2IBIS as all our netlists and simulations decks are in spectre. One more thing is, will S2IBIS version 1.3 supports new features of IBIS like 4.2 version of IBIS ? Thanks for your help! Best Regards, Sudarshan HN NXP Semiconductors/CTO /PLT # 1 Murphy Road, Ulsoor, Bangalore - 560 008 , India. Ph:+91-80-40247073 Fax: +91-80-4024 7855 seri:sudarsha@inpsblr E-mail: sudarshan.honnudike@nxp.com "Muranyi, Arpad" <arpad.muranyi@intel.com> Sent by: owner-ibis-users@server.eda.org 2006-11-08 10:13 PM To <ibis-info@server.eda.org> <ibis-users@server.eda.org> cc Subject RE: [IBIS-Users] Warnings in V-t curves Classification Sudarshan, One thing that I haven't seen in any of the responses yet is the definition of Vmeas. From the IBIS specification: | The Vmeas subparameter is the timing reference voltage level | that the semiconductor vendor uses for the model. Include | Cref, Rref, Vref, and Vmeas information to facilitate | board-level timing simulation. In other words, this is the voltage at which Tco is measured with the load circuit which is described by Rref, Cref, Vref. This information is usually available in the data sheet of a chip. You need to have an understanding of what Tco means, how it is measured for the data sheet, and what it is used for in the timing spreadsheet together with the flight time and setup and hold time parameters. Welcome to signal integrity! If you have a buffer that cannot drive through this value using the give Vref, Rref and Cref load, Vmeas becomes useless. Arpad =============================================================== _____ From: owner-ibis-users@server.eda.org [mailto:owner-ibis-users@server.eda.org] On Behalf Of Sudarshan Honnudike Sent: Wednesday, November 08, 2006 12:22 AM To: Lynne D. Green; Bob Ross Cc: ibis-info@server.eda.org; ibis-users@server.eda.org Subject: RE: [IBIS-Users] Warnings in V-t curves Hi Bob & Lynne, I want to get some more clarifications on your answers. Lynne told that , Vmeas should be between Vinl and Vinh. As Vin = 1.2 V , Vinl and Vinh would be 0.36 and 0.84 V respectively . In that case Vmeas should be between 0.36 and 0.84. But if we connect R_fix to VDDE(2.5 V) then in falling waveform, it will fall to 1.25V. So in that case it will not cross Vmeas. So according to what Bob suggested, Vmeas should be between 1.25 and 2.5 V. Please correct me if i am wrong in explanation. And as Bob said, it is an open drain cell. I have taken R_fix as 44 ohm. I have changed R_ref to 50 ohm , Vmeas to 1.25 V and V_ref to 1.3V. Now I am able to get rid of those warnings. From the cookbook I understood that Vref, Cref, Rref and Vmeas are AC test load conditions. i.e, under this setup, buffer's delay information is measured. Can we use the same value of R_fix, C_fix and V_fix to Rref , Cref and Vref parameters, as under these parameters we measure the V-t waveforms. ? Please let me know your answers. Best Regards, Sudarshan HN NXP Semiconductors/CTO /PLT # 1 Murphy Road, Ulsoor, Bangalore - 560 008 , India. Ph:+91-80-40247073 Fax: +91-80-4024 7855 seri:sudarsha@inpsblr E-mail: sudarshan.honnudike@nxp.com "Lynne D. Green" <lgreen22@mindspring.com> 2006-11-08 12:22 PM To "'Sudarshan Honnudike'" <sudarshan.honnudike@nxp.com> <ibis-users@eda.org> <ibis-info@eda.org> cc Subject RE: [IBIS-Users] Warnings in V-t curves Classification Hello, Sudarshan, 1) Please use ibischk4. This is capable of checking [IBIS ver] 3.2 models. Bug fixes and enhancements are not patched back into older versions of ibischk. 2) Rref, Cref, Vref and Vmeas are used for hardware testing, and should be on the I/O data sheet or on the I/O interface specification. Vmeas is normally between Vinl and Vinh. 3) Rref default value is 50 ohms. 1E+9 is not a reasonable value. 4) Vcc_core is not used in the IBIS model. It is used only in simulation to drive the "core" pins (Enable, Data) on the I/O buffer. 5) There are two good sources of modeling help. These are the IBIS Cookbook http://www.eigroup.org/ibis/tools.htm and the IBIS Quality Checklist http://www.vhdl.org/pub/ibis/quality_wip/ Best regards, Lynne "IBIS training when you need it, where you need it." Dr. Lynne Green Green Streak Programs http://www.greenstreakprograms.com 425-788-0412 lgreen22@mindspring.com- Lynne ________________________________ From: owner-ibis-users@eda.org [mailto:owner-ibis-users@eda.org] On Behalf Of Sudarshan Honnudike Sent: Tuesday, November 07, 2006 9:25 PM To: ibis-users@eda.org; ibis-info@eda.org Subject: [IBIS-Users] Warnings in V-t curves Hi Experts, I am getting the following warnings while running ibischk3 on the buffer models. WARNING - Model 'bsppon3mchpt3v3_p': TYP VI curves cannot drive through Vmeas=0.6V given load Rref=1e+09 Ohms to Vref=0V WARNING - Model 'bsppon3mchpt3v3_p': MIN VI curves cannot drive through Vmeas=0.6V given load Rref=1e+09 Ohms to Vref=0V WARNING - Model 'bsppon3mchpt3v3_p': MAX VI curves cannot drive through Vmeas=0.6V given load Rref=1e+09 Ohms to Vref=0V Our buffer models will be operating at 2 voltages coreside voltage and pad side voltage. In this case, coreside voltage(VDD) is 1.2 v and pad side voltage(VDDE) is 2.5v. So in V-t waveforms 1. Rising waveform with R_fix connected to VDDE(2.5v): waveforms are varying from 1.249v to 2.5v 2. Falling waveform with R_fix connected to VDDE(2.5v):waveforms are varying from 2.5v to 1.254v I have a confusion regarding the selection of Vmeas. Should it be half of corevoltage(1.2/2) or half of pad voltage (2.5/2)? Please let me know how this parameter is used in the application (may be in SI or somewhere else). I even changed Vmeas to 1.25 V. Still I am getting warnings like WARNING - Model 'bsppon3mchpt3v3_p': TYP VI curves cannot drive through Vmeas=1.25V given load Rref=1e+009 Ohms to Vref=0V WARNING - Model 'bsppon3mchpt3v3_p': MIN VI curves cannot drive through Vmeas=1.25V given load Rref=1e+009 Ohms to Vref=0V WARNING - Model 'bsppon3mchpt3v3_p': MAX VI curves cannot drive through Vmeas=1.25V given load Rref=1e+009 Ohms to Vref=0V Please let me know your answers. Best Regards, Sudarshan HN NXP Semiconductors/CTO /PLT # 1 Murphy Road, Ulsoor, Bangalore - 560 008 , India. Ph:+91-80-40247073 Fax: +91-80-4024 7855 seri:sudarsha@inpsblr E-mail: sudarshan.honnudike@nxp.com -------------------------------------------------------------------- |For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org |with the appropriate command message(s) in the body: | | help | subscribe ibis <optional e-mail address, if different> | subscribe ibis-users <optional e-mail address, if different> | unsubscribe ibis <optional e-mail address, if different> | unsubscribe ibis-users <optional e-mail address, if different> | |or e-mail a request to ibis-request@eda-stds.org. | |IBIS reflector archives exist under: | | http://www.eda-stds.org/pub/ibis/email_archive/ Recent | http://www.eda-stds.org/pub/ibis/users_archive/ Recent | http://www.eda-stds.org/pub/ibis/email/ E-mail since 1993Received on Fri, 10 Nov 2006 11:12:46 -0500
This archive was generated by hypermail 2.1.8 : Fri Nov 10 2006 - 08:14:18 PST