Does anybody have a snap_tut_spec.fpg built with mlib_devel branch m2021a that works with casperfpga branch py38-dev, and passes the rampTest calibration procedure? I'm still having issues.
If I use the older branch py38, and the example .fpg file included in tutorials_devel at https://github.com/casper-astro/tutorials_devel/tree/main/snap/tut_spec , I do get it to work. It calibrates using rampTest. Annoyingly, fpga.adcs['snap_adc'] was not automatically populated after fpga.upload_to_ram_and_program(), meaning I had to create an adc object "by hand" using" adc=casperfpga.snapadc.SnapAdc(fpga, num_chans=numChannel, resolution=8, ref=10). The py38 branch does NOT work for the .fpg file that I jasper build from the snap_tut_spec.slx. If I try to fpga.upload_to_ram_and_program() that .fpg build on py38, I get an error in CasperFpga._create_casper_adc_devices. If I use the recommended branch py38-dev, I CANNOT load the example .fpg file included in tutorials_devel. It does not populate fpga.adcs with this file, nor can I build the SnapAdc object "by hand" like I did above. I can, however, load my own .fpg build, and it does properly create adc=fpga.adcs['snap_adc']. Unfortunately, calling adc.alignFrameClock() runs forever. This is called by adc.rampTest() which therefore also runs forever. This is what happens with I use the example adc calibration code developed by Wei Liu, which you linked to above. Since I cannot re-build snap_tut_spec.slx, load it, and calibrate, I can't make any changes. I'm having similar issues with the correlator tutorial. The stock file snap_tut_corr.fpg loads and calibrates under py38, but it returns nonsense spectra (half zero) and doesn't work in py38-dev. I can't add or tweak things in the .slx because then it hangs on the rampTest calibration like the tut_spec examples. Thanks, Jason On Tuesday, May 16, 2023 at 3:23:26 AM UTC+10 Jonathon Kocz wrote: > >> Wei Liu's code helped because it gave me the idea to call rampTest() >> first. Is rampTest() the only "calibration" that Wei Liu's code is >> doing? rampTest() calls alignLineClock() and alignFrameClock(). Is this all >> that you mean by "calibration"? Not some fine-tuning or balancing of gains >> or delays for different sub-ADCs, which then get interleaved together? >> >> As you probably know the ADC has 8-cores (each of the 4 main cores are > actually two sub-cores which are interleaved, and they start in a random > order on power up. The main focus of the casper calibration code is for > ordering these samples in the correct way. Aligning the clocks should also > deal with the delays, but we don't apply gain inter-core corrections. > >> >> adc.set_gain() still does nothing, even when called where Wei Liu's code >> calls it. Where in the startup process should this get called? >> > > I believe that set gain is for input signals, so it won't impact the > actual calibration (I would actually recommend keeping it below 8, it moves > into a "digital" gain after that which mostly just confuses things). > > The py38-dev and py38 branches of casperfpga differ in only trivial ways >> in snapadc.py. Maybe the changes you spoke of have already been merged into >> py38? >> >> py38 was supposed to work, but there were still some hiccups - Wei fixed > them up, so they should be close. > > >> As for the build error, I've only tried SNAP tutorial 1 and 3 and I got >> the error for both. Once Xilinx finally gets back to me about an academic >> license, I'll be able to try it on my RFSoC 4x2. Does it normally take more >> than a month to process a donation request? I've asked twice. >> >> For a software donation, I've typically found it to be fairly fast > (almost automated - a few days at most), but it has been a long time since > I last asked - someone with more recent experience might be able to chime > in. > > Best, > Jonathon > > > >> -- You received this message because you are subscribed to the Google Groups "[email protected]" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/c8fbfe42-2a41-4e41-b11f-376e25776d02n%40lists.berkeley.edu.
