How to extract the compressed timesample?

[costineesti]

Hi! I've been trying to parse an MF4 file (4.1) to extract the total simulation time. First, I go to addres 0x40 (HD) in the file and then simply use this structure to get further and further:

(
next_cg_addr, first_ch_addr, acq_name_addr, acq_source_addr, first_sample_reduction_addr, comment_addr, cg_master_addr, record_id, cycles_nr, flags, path_separator, reserved1, samples_byte_nr, invalidation_bytes_nr,
) = v4c.CHANNEL_GROUP_RM_SHORT_uf(stream, address + COMMON_SIZE)

So, I don't use the library as I want to manually parse it.
I saw that cycles_nr gives me the simulation time but multiplied by a factor: (200.175 seconds are 200175 in cycles_nr). What I want now is to find the address where I can find that 0.001 factor to be able to multiply it with cycles_nr and get the true simulation time. Can you please help? I can provide code if necessary. I've been following these schemas:

[costineesti]

@danielhrisca

[danielhrisca]

The conversion rule is linked to the channel metadata

[costineesti]

@danielhrisca , apparently there is no conversion rule in my case. I found these ##DZ blocks and maybe decompressing them could lead to the original timestamps? I see no other way

[danielhrisca]

This means that the raw values are actual the timestamp values in float format. See the record format description from the MDF wiki https://www.asam.net/standards/detail/mdf/wiki/ on how to extract the time channel samples after you decompress the DZblock

[costineesti]

@danielhrisca , I've successfully decompressed the first DZ block, but when I save the HEX Dump file, I can't retrieve the raw byte values correctly. How can I ensure that the bytes remain in the correct order after decompression? I applied the transpose rule, but it did not resolve the issue.

In the software I can see the original timestamp:

But after I decompress, I get this:

For some context, this is what I did in code: