According to the documentation (spectre -h relay), this component supports the "isnoisy" parameter. However, it cannot be set from the component's properties (see screenshot below).
As you can see, I tried manually adding it as a user string property, but it is being ignored by the netlister (Spectre).
How can I control this flag?
Thanks and regards, Jorge.
ICADVM 18.1.500.6 / Spectre 18.1.0.421.isr9
Hi,
I have searched this problem a lot but couldnt find anywhere. Therefore I had to ask it here. If anyone knows a related question please send me the link.
I am working on a project and doing transient simulations using ADE L. There are multiple output nodes to be saved but I just wanted to update my save list and I went and selected
'Output>ToBeSaved>SelectOnDesign' in ADE L window.
When I come back to schematic screen to save my new nodes, I saw my already saved nodes probed. And I just clicked new nodes which need to be saved but it didnt probe them and still expected me to select a node although I clicked that node multiple times. I checked Virtuoso Log screen but couldn't see any related message about the issue.
I am using Virtuoso 6.1.6. -64b
Any help would be more than appreciated...
Hello
While performing LVS check for a device i am getting following error messages
1. malformed device
2. The layout cell is bound to multiple schematic cells.
I already checked the device parameters (w, l , m, no of gates) between schematic and layout all are same plus the layout is not linked to more than one schematics. I am not sure why getting these LVS errors.
Thank you .
Hi,
this question was asked often in the forum (in different ways), but I did not yet find a definitive answer for this:
I am using Assembler to perform simulations.
I am doing a DC sweep on variable "Vgs". The variable is swept by the DC simulation from 0 to 1.0 - not via the tests DesignVariable.
The test DesignDariable is set to "0.99".
Now I want to use the swept variable in output equations via VAR("vgs") - but this gives me a constant of 0.99, and not the expected swept values.
Why is that and how can I access the swept variable ?
Of course I can do the sweep using the DesignVariable "vgs 0:0.1:1", but this creates 11 simulations, that take much longer.
The actual circuit is complex and I want to save simulation time.
I am using IC6.1.7.1400.500.
Please let me know what the best way is to access the sweep variable values.
Dear All,
I want to generate a clock input which is going to be triggered by another signal in the testbench.
The vsource can give a clock but here we can add a fix delay only before the simulation starts.
I used a verilogA block using timer to generate the clock. But its frequency is changing in simulation unlike vsource clock.
Similarly, idtmod in VerilogA is also giving inaccurate results.
Can anybody please tell how to add an accurate triggered clock in a testbench.
Kind Regards,.
Dear All,
After moving to IC6.1.8-64b.500.7, my previously functioning MC scripts became more finicky about history entries. Admittedly this previously working script had a lot of hacks in it.
I have a loop where I want to overwrite the previous run's history and keep the same history name.
The error that I am stuck on is:
ERROR (EXPLORER-2406): A history entry named uncal.0 cannot be created.
There is already a history entry with that name.
Delete the existing entry or choose a different name.
My script has the following:
;ocnxlSetOverwriteHistory( t )
;ocnxlSetOverwriteHistoryName( "uncal.0" )
;ocnxlRenameCurrentHistory("old")
ocnxlHistoryPrefix( "uncal" )
ocnxlRun( ?mode 'monteCarlo ?nominalCornerEnabled t ?allCornersEnabled t ?allSweepsEnabled t)
The commented out lines were commands that I tried to overcome the error. Is there a delete history command? I figured that the two OverWrite commands would have worked.
The big picture is that I have a loop where a Monte Carlo simulation runs in order to calculate a correction, and then the simulation is rerun with the same seed and the correction applied. During each iteration I do not want to accumulate data (psf files) after the outputs have been collected, so overwriting or deleting after each run would be sufficient.
Best regards,
Paul
Dear All,
In the past I would interrupt an ocean script by issuing a ctrl-c, and this would stop the present simulation. I am finding that the behaviour is not the same with ocean XL(IC6.1.8-64b.500.7), and there is no response to ctrl-c.
I start ocean from the command line with just "ocean", and begin the script with either:
ocnSetXLMode("assembler") or ocnSetXLMode("explorer")
The simulation is a harmonic balance simulation, with APS selected and 16 threads.
To interrupt the script I find that I need to seek out the spectre PID and use kill -9.
Is there a more interactive setup that I can use that will respond to an interrupt, or am I using the best practice?
Best regards,
Paul
Hello,
I am trying to use pole zero anayses (PZ) using ADE spectra to locate the poles and zero of my amplifier, but I think I am missing the correct setting, if there is any tutroial or help to expalin it more it will be very useful to me
Thank you very much
Hello
I read a couple of papers in which the author performed a montecarlo based on geometric (dimention) variations of the elements in the circuit. is there any setting in MC for that ?
If you like to have a look here are the papers
ieeexplore.ieee.org/.../4349224
link.springer.com/.../s10470-016-0874-2
Also in the first paper the author run the Montecarlo in over process corners (SS, FF...etc), why he should include these corners when it will already covered by the Montecarlo variation ?
Usually for me I run with MC two other corners that are supply voltage and temperature variation but never tried to include process corners with it nor having an idea about including geometric variation
Thank you very much
how we can calculate the rise time and fall time in cadence software(6.1.5). what are the expression for rise time and fall time?
Hi,
I exported a schematic with multiple hierarchy levels to html from schematic editor xl, and the resulting file is perfectly visible in firefox under linux.
The same file is not rendered under windows (I tried firefox, edge, IE11); the browser hangs and at most I see a large collection of text.
Has anyone experienced the same problem?
Greetings!
I am simulating a schematic with ADEXL. This circuit has a trimming to compensate process and mismatch variations. I have a test where I find the best trimming word and other to measure the circuit response at this trimming word. I've done the test dependency with calcVal() and for corner simulation it is working just fine.
The problem is when I run monte carlo. It looks like it takes different samples for each test, in a way that the word found in the first test is not useful to improve the performance of the second test.
Any ideas?
Thank you!
I am running HB analysis and notice that I can plot the THD from the direct plots form, but I would like to use that value for an optimization run, so I need to access it in the "Outputs Setup" tab as a function expression. Is that possible? If not I would like to use the calculator to manually calculate the THD, I am using the "vfreq" and "conjugate" functions to calculate the square of each coefficient but I can't figure out how to add them up. Is there a way to add the coefficients up; or more generally, can I add a signals amplitude without taking the integral(multiplying by dt)?
I'm getting the error when I'm trying to create schematic on Schematic editor. it shows an error like .*Error* hiCreateCyclicField: value must be contained within the list of choices".i m using virtuoso 6.1.5 version.
Hi
When I am trying to export gds from virtuoso.
The streamout form is not opening. I faced this issue for a couple of times now.
I tried reopening the virtuoso but is has not worked.
it is executing the command pipoDisplay(transStreamOutForm) But its not opening.
please help me.
Thanks,
Ganesh Doddipatla.
Hi All,
Is there a free cadence Host Id or Reference Key?, so that
I can access Cadence Online Support web site, because my entitlement
has expired.
Best regards,
Marben
Dear All,
I have to run a multi-corner ADE-XL simulation and I have to save the waveforms into separate files.
Later I can use these files in pwl source for other test-benches. This way I can fasten the simulation for later cases.
But, I am wondering how I can save the waveform (NOT manually) into files through ADE-XL run. Can it be done through SKILL script.
Could anybody please help me in this regard.
Kind Regards,
Hello,
Reference link: https://community.cadence.com/cadence_technology_forums/f/custom-ic-design/40093/matlab-cds_srr-dc-sweep-of-mosfet-parameters
I got the simulation in the above link set up to work for dc parametric simulation where I can read the data into Matlab using cds_srr Matlab specter function.
Then I did a noise simulation in ADE L and I see that the noise data is stored in ADE L under the result browser named noise. For noise analysis in ADE L, I set my sweep to Logarithmic and points per decade to 100. I probed an ideal voltage source with a DC value of zero. In the noise Analysis option, under Output Noise, I selected probe and for output probe instance, I selected the Ideal voltage source with the DC value of zero (I probed the ideal voltage source). For input noise, I selected voltage and selected the gate-source voltage (vgs) because I want to refer the output noise to the input where I want to compare the noise signal to the input signal.
How do I get to read the noise data in Matlab via cds_innersrr function since cds_crr is buggy for noise? Where is the noise being stored in the psf? Because for my output file this is all I see this:
cds_innersrr(c.outfile )
PSFversion: 1.00
dcOp-dc (dc).
dcOpInfo-info (info).
dc-dc (dc).
modelParameter-info (info).
element-info (info).
outputParameter-info (info).
designParamVals-info (info).
primitives-info.primitives (info.primitives).
subckts-info.subckts (info.subckts).
variables (design_variables).
dcOpViolations-violations (violations-dc).
dcViolations-violations (violations-dc).
ans =
12×1 cell array
{'dcOp-dc' }
{'dcOpInfo-info' }
{'dc-dc' }
{'modelParameter-info' }
{'element-info' }
{'outputParameter-info' }
{'designParamVals-info' }
{'primitives-info.primitives'}
{'subckts-info.subckts' }
{'variables' }
{'dcOpViolations-violations' }
{'dcViolations-violations' }
Thank you very much in advance regarding this.
I'm having trouble getting the MPT Coloring to work. I'm sure I'm just missing something simple.
I have run (mptActivate t) but in the Properties window for my rectangle the MPT section remains grayed out. I am using the same techlib, which includes colorMap, and same oalibs as other colleagues who can use the properties but were unfortunately unable to tell me how to fix mine. Are there other settings I need to enable?
Hello All,
In Virtuoso schematic how do I setup up "rules check" to give error during netlist out in case of missing CDF param ?
ICADV12.3-64B
Thanks.