Sniffing the Rigol's internal I2C bus

[tlu]

... I'm curious as to see what you mean when you say options are not labeled.

Under the Options menu, the 5 items are simply numbered 1 to 5 without any labels indicating what each represents.  I know item #1 is the Tracking Generator option because that's the only option enabled on my unit (I wasn't lucky enough to get any trial options  ).  Earlier in this thread (#794) Spark posted a picture of a box label listing the included options in a Yes/No format (just like the option screen) with the TG also as the top option.  If we correlate that sequence to the license screen, the option numbers are:

1) Tracking Generator
2) Advanced Measurement Kit
3) 10 Hz RBW
4) EMI/Quasi Peak
5) VSWR

The only options offered so far are 1,2,4, & 5.  Since 5 options are listed on all the 815's, the 5th option they provide space for (presumably option #3) is for the 10 Hz RBW and is available on all 815's with the correct key.  Out of the 5 options, 10 Hz RBW is the one option I probably would have paid for if they had made it available  .

Thanks for clearing that up Marc. I'm pretty optimistic now that the remaining option would be the 10hz RBW and it pre-built into the dsa815.

Has anyone take a crack at an attempt to see if this option is hidden somewhere in the firmware? It's probably a matter of time before someone will figure out how to enable this option. Can't wait for mine to arrive.

[nedos]

Speaking of firmware, I see that Rigols website now lists the latest firmware, as of July 31, for the DS4000 to be 00.02.00 when it just a month ago was 00.01.00.07 or something like that (I'm currently running neither of those). I wonder if they have already implemented a fix for the hack....anyone knows?

I just upgraded to 02.00. Everything still works. I entered my option key after upgrading to 02.00.

Thanks for the hard work, guys!

[synapsis]

Hi,
Has anyone tried using the protocol decoders?
My DS4000 series came without any trial licenses (at the time I didn't even know there WAS trial licenses to be had) so I never got to actually try it before installing the "license". Last night I tried the RS232 decoder and it's so PAINFULLY slow it's almost unusable. I don't know if I'm doing it wrong or what (don't think so) but geez, right now I'm SO glad I didn't pay for it.

I wonder if it's a firmware thing.... I'd appreciate if anyone else could report your experience using the RS232 (or any other) decoder, preferably on a DS4000 series (but 2000 would be interesting too) and what firmware version you're running, thanks!

Speaking of firmware, I see that Rigols website now lists the latest firmware, as of July 31, for the DS4000 to be 00.02.00 when it just a month ago was 00.01.00.07 or something like that (I'm currently running neither of those). I wonder if they have already implemented a fix for the hack....anyone knows?

Thanks!

I've used the RS232 decoder on a DS2000. The only issue I had was that using the preset 57600 baud didn't work correctly, so I went into the user baud rate setting and tweaked the rate on a known packet of data. This was before I upgraded to 05.

Other than that, it didn't have any missing bytes and it ran at whatever timebase I used. The table even updated faster than I could read it.

[alank2]

I've used the RS232 decoder on a DS2000. The only issue I had was that using the preset 57600 baud didn't work correctly, so I went into the user baud rate setting and tweaked the rate on a known packet of data. This was before I upgraded to 05.
Other than that, it didn't have any missing bytes and it ran at whatever timebase I used. The table even updated faster than I could read it.

It seemed like the decoder would only decode from what is visible on the screen and not from the trigger point, has anyone else experienced this?

[H.O]

Thanks guys,
In my case I send a string "Test x" where x "counts" from A to Z. I'm sending at 9600baud with 100ms delay between packets. Pretty relaxed if you ask me.

On the scope, set to 1ms/div, triggering on falling edge, I can see the trace and the databits in the last byte "counting", however the decoded data shows A, O, C, Q, E, S, G, U, I, W... as the last byte which tells me there's 1300-1400ms between updates of the decoded data on the screen. And when the decoded data actually IS updated the "sweep" freezes for hundreds of ms. Changing the time/div from 1ms/div to 500us makes it even slower.

Changing the memory depth from Auto to 7kpts (lowering the sample rarte from 2GSa/s to 500k) makes it "a lot" faster but it's still "skipping" (A, D, G, J, M, P...) and now the triggering becomes unstable (which might be "fixable" with some holdoff - didn't try).

If I'm not doing it wrong I'm going to call it on the edge of useless, I really do hope I AM doing something wrong OR that a newer firmware fixes it.

Anyone else with a DS4000 out there that care to try it?

Thanks again!

[jasonbrent]

According to tequipment, my DS2072 (replacing a 3 day old siglent 1102cml that died(!)) should ship on 8/16. I'm hoping it's hardware version 2.0 and that I can squeak 100Mhz+ out of it using the info in this thread.

-jbl

[fake-name]

Hi,
Has anyone tried using the protocol decoders?
My DS4000 series came without any trial licenses (at the time I didn't even know there WAS trial licenses to be had) so I never got to actually try it before installing the "license". Last night I tried the RS232 decoder and it's so PAINFULLY slow it's almost unusable. I don't know if I'm doing it wrong or what (don't think so) but geez, right now I'm SO glad I didn't pay for it.

I wonder if it's a firmware thing.... I'd appreciate if anyone else could report your experience using the RS232 (or any other) decoder, preferably on a DS4000 series (but 2000 would be interesting too) and what firmware version you're running, thanks!

Speaking of firmware, I see that Rigols website now lists the latest firmware, as of July 31, for the DS4000 to be 00.02.00 when it just a month ago was 00.01.00.07 or something like that (I'm currently running neither of those). I wonder if they have already implemented a fix for the hack....anyone knows?

Thanks!

The decoders on my 4000 series were pretty laggy before I used the trial keys. They don't seem to have changed after installing the key.

[DL5TOR]

some good News for the DSA815:

The licensekey uses the same algo and keys then the DS2000 but the Serialnumber used is somthing different

also intresting is that i found the srings DSA830A DSA830 DSA820 and DSA815 in there
I am still in the porces of finding the rest

i hope i find somthing
73

[BravoV]

This thread is getting better and better everyday !   

[olsenn]

some good News for the DSA815:

The licensekey uses the same algo and keys then the DS2000 but the Serialnumber used is somthing different

also intresting is that i found the srings DSA830A DSA830 DSA820 and DSA815 in there
I am still in the porces of finding the rest

i hope i find somthing

You are my hero! Please keep all the goods on this forum thread so we can all follow along. All our base are belong to you!

[jsykes]

some good News for the DSA815:

The licensekey uses the same algo and keys then the DS2000 but the Serialnumber used is somthing different

also intresting is that i found the srings DSA830A DSA830 DSA820 and DSA815 in there
I am still in the porces of finding the rest

i hope i find somthing

You are my hero! Please keep all the goods on this forum thread so we can all follow along. All our base are belong to you!

 
I second this.
In view of the Rigol model number system format it would seem that the DSA820 would be a 2GHz model and the 830A a 3GHz model with 10Hz RBW but that doesn't seem to make sense since there would seemingly be no difference between those and some of the DSA1000 models. I may be wrong, but I believe the DSA815 was released after the DSA1000 series. Maybe they were being designed at the same time with the 1000's being released first and the thought of 820 and 830 models was dropped in favor of the 1000 models...just a guess.
Thanks for everyone's hard work. 10Hz RBW on the DSA815 would especially be a godsend for us.

[geostep]

According to tequipment, my DS2072 (replacing a 3 day old siglent 1102cml that died(!)) should ship on 8/16. I'm hoping it's hardware version 2.0 and that I can squeak 100Mhz+ out of it using the info in this thread.

-jbl

My DS2072 was delivered from tequipment last Thursday the 8th.  Mine came with hardware rev 1.0.1.

The DS2072 started having problems right away.    The front panel controls started to randomly freeze.  All buttons, knobs and lights became unresponsive.  It's done it several times in the last few days.  Although the display continued to work worked fine and displayed signals no problem.  The only way to unfreeze it was to reboot.   

I updated the firmware from 00.01.00.00.03  to  00.01.01.00.02  but that did not help.  Also the select knob was rather touchy and often would overshoot the desired setting by a few, erm, settings.

Needless to say I'm going to have to call tequipment on Monday and see about getting it replaced. *sigh*

- George

[tlu]

some good News for the DSA815:

The licensekey uses the same algo and keys then the DS2000 but the Serialnumber used is somthing different

also intresting is that i found the srings DSA830A DSA830 DSA820 and DSA815 in there
I am still in the porces of finding the rest

i hope i find somthing
73

Nice find. Hopefully, this will manifest into a 3Ghz BW and 10Hz RBW. Keep up the good work.

[warp_foo]

With 57 pages on this topic, has anyone summarized what the current state and process of the DS2000 hack is? I'd like to buy a DS2072, but before pulling the trigger, I'd like to make sure I understand what's been discussed here. 

Thanks,

m

[true]

According to tequipment, my DS2072 (replacing a 3 day old siglent 1102cml that died(!)) should ship on 8/16. I'm hoping it's hardware version 2.0 and that I can squeak 100Mhz+ out of it using the info in this thread.

-jbl

My DS2072 was delivered from tequipment last Thursday the 8th.  Mine came with hardware rev 1.0.1.

The DS2072 started having problems right away.    The front panel controls started to randomly freeze.  All buttons, knobs and lights became unresponsive.  It's done it several times in the last few days.  Although the display continued to work worked fine and displayed signals no problem.  The only way to unfreeze it was to reboot.   

I updated the firmware from 00.01.00.00.03  to  00.01.01.00.02  but that did not help.  Also the select knob was rather touchy and often would overshoot the desired setting by a few, erm, settings.

Needless to say I'm going to have to call tequipment on Monday and see about getting it replaced. *sigh*

- George

My select knob was a bit problematic, I just worked it a bit and now it's fine.

[jasonbrent]

With 57 pages on this topic, has anyone summarized what the current state and process of the DS2000 hack is? I'd like to buy a DS2072, but before pulling the trigger, I'd like to make sure I understand what's been discussed here. 

Thanks,

m

^^ This. I ordered a DS2072 that should ship this coming week... My impression from the thread is that I can turn it into a fully-featured ~200Mhz scope with what was learned in this thread. Or at least a 70Mhz scope with all features enabled. I'm hoping it's the former rather than the latter. :-)

-jbl

[Chalky]

With 57 pages on this topic, has anyone summarized what the current state and process of the DS2000 hack is? I'd like to buy a DS2072, but before pulling the trigger, I'd like to make sure I understand what's been discussed here. 

Thanks,

m

^^ This. I ordered a DS2072 that should ship this coming week... My impression from the thread is that I can turn it into a fully-featured ~200Mhz scope with what was learned in this thread. Or at least a 70Mhz scope with all features enabled. I'm hoping it's the former rather than the latter. :-)

-jbl

The former! 

[jasonbrent]

Yay. :-)   

[warp_foo]

With 57 pages on this topic, has anyone summarized what the current state and process of the DS2000 hack is? I'd like to buy a DS2072, but before pulling the trigger, I'd like to make sure I understand what's been discussed here. 

Thanks,

m

^^ This. I ordered a DS2072 that should ship this coming week... My impression from the thread is that I can turn it into a fully-featured ~200Mhz scope with what was learned in this thread. Or at least a 70Mhz scope with all features enabled. I'm hoping it's the former rather than the latter. :-)

-jbl

The former! 

Awesome. Is there a summary of 'what was learned in this thread' with the relevant bits and the process the hack follows to make a 2072 into a 2202?

Thanks,

m

[tlu]

Getting back to the dsa815, can the 10Hz RBW be enabled with the other trial options? Just curious as no one has mentioned of being able to enable it since it shows up as an unlabeled option.

[synapsis]

With 57 pages on this topic, has anyone summarized what the current state and process of the DS2000 hack is? I'd like to buy a DS2072, but before pulling the trigger, I'd like to make sure I understand what's been discussed here. 

Thanks,

m

^^ This. I ordered a DS2072 that should ship this coming week... My impression from the thread is that I can turn it into a fully-featured ~200Mhz scope with what was learned in this thread. Or at least a 70Mhz scope with all features enabled. I'm hoping it's the former rather than the latter. :-)

-jbl

The former! 

Awesome. Is there a summary of 'what was learned in this thread' with the relevant bits and the process the hack follows to make a 2072 into a 2202?

Thanks,

m

Reply #572 has (one of the) Linux versions of the keygen, reply #574 has the Windows version. I think it's page 39ish with default forum settings.

[davidefa]

Thanks guys,
In my case I send a string "Test x" where x "counts" from A to Z. I'm sending at 9600baud with 100ms delay between packets. Pretty relaxed if you ask me.

On the scope, set to 1ms/div, triggering on falling edge, I can see the trace and the databits in the last byte "counting", however the decoded data shows A, O, C, Q, E, S, G, U, I, W... as the last byte which tells me there's 1300-1400ms between updates of the decoded data on the screen. And when the decoded data actually IS updated the "sweep" freezes for hundreds of ms. Changing the time/div from 1ms/div to 500us makes it even slower.

Changing the memory depth from Auto to 7kpts (lowering the sample rarte from 2GSa/s to 500k) makes it "a lot" faster but it's still "skipping" (A, D, G, J, M, P...) and now the triggering becomes unstable (which might be "fixable" with some holdoff - didn't try).

If I'm not doing it wrong I'm going to call it on the edge of useless, I really do hope I AM doing something wrong OR that a newer firmware fixes it.

Anyone else with a DS4000 out there that care to try it?

Thanks again!

I have a DS2072, I have duplicated your setup ( a pc sending 'Test X' at 9600 baud with a 100mS delay between strings ):
- 1mS/div, 1MSa/S, 14K memory
- trigger normal ( I can't get a stable triggering in auto, quite annoying )
- serial decoder enabled

I can see the decoded bus 'changing' on the fly, but I can't see the sequence A B C... ( the letters are changing 10 times a second )
I connect a second scope on the trigger output and I can see a 10Hz stable signal ( so the scope isn't missing any string )
If I record a few seconds of signal ( this is an easy and great feature af this scope, isn't it? ) and later review the recording a can see clearly the sequence of A B C... with no missing letters
Hope this helps

P.S. thanks to everyone for this great tread

[PA0PBZ]

The decoding seems a lot slower than the waveform update speed, if you set your timebase slow enough and send a long string of RS232 characters you can see it building the decode. Not really a problem I think, you are not going to read along real time anyway, at least I'm not. I tested some other stuff too, and found that you can really fill the memory and then 'browse' through and it will decode, but you have to stop the auto triggering. If it is in auto trigger mode you can't even scroll the data: the waveform will scroll but the decoded output will stay in place. And it even shows the decoded data in zoom mode!

[H.O]

Thank you,
To clarify, on my DS4000, sending 10 packets/strings per second, the waveform updates properly 10 times per second as expected (running in Normal mode) - EXCEPT when the decoded data is actually updated every 1300-1400ms or so - at that instant the waveform frezzes for a couple of 100ms. I have not checked the trigger output to see if the aquisition stops or if it's just the screen update.

I can capture long events at slower timebase and then scroll thru it, zooming in or use waveform record to capture multiple packets and then scroll thru and decode them, no problems, I just find the actual decoding process painfully slow.

I obviously expect it to fall apart at some point but really, 1300-1400ms to decode a string of 6 chars and display the data seems awfully slow. If it's not decoding the data in "real time" how will it possible to trig on a specific char/byte, something I haven't tried yet?

I know the Agilent 3000x series scope do the decoding in hardware, something Rigol clearly isn't doing so I don't expect the performance to be like what can be seen in this demonstration but IMHO the Rigol is REALLY slow.

Thanks again!

[PA0PBZ]

If it's not decoding the data in "real time" how will it possible to trig on a specific char/byte, something I haven't tried yet?

Is that possible? I have not seen a menu option to do that (DS2000)