Author Topic: SPI Decoding  (Read 12912 times)

0 Members and 1 Guest are viewing this topic.

Offline echen1024Topic starter

  • Super Contributor
  • ***
  • Posts: 1660
  • Country: us
  • 15 yo Future EE
SPI Decoding
« on: November 02, 2013, 04:13:17 am »
When browsing through the options on the Rigol DS2072, the serial decoding package includes SPI. How does it decode SPI with only 2 channels? Does it possibly only decode MOSI and MISO one at a time? Or does it use the ext. trigger. I am still a bit torn between this scope and the DS1000Z.
I'm not saying we should kill all stupid people. I'm just saying that we should remove all product safety labels and let natural selection do its work.

https://www.youtube.com/user/echen1024
 

Offline Mark_O

  • Frequent Contributor
  • **
  • Posts: 939
  • Country: us
Re: SPI Decoding
« Reply #1 on: November 02, 2013, 04:35:56 am »
When browsing through the options on the Rigol DS2072, the serial decoding package includes SPI. How does it decode SPI with only 2 channels?
It doesn't.  Really.   :wtf:

Quote
Does it possibly only decode MOSI and MISO one at a time?
That's all it  can do.  Half the conversation .  And even that not completely (no Device Select gating possible).  See my comments here.

Quote
Or does it use the ext. trigger.
Sadly, no.

As I commented farther down in the same thread...

Quote from: Mark_O
That's what's going on with SPI on the DS2000.  You only get to see half the conversation at one time, because one of your two available channels HAS to be used for the Clock.  On a SPI bus, the CPOL and CPHA are critical (polarity and phase), because they determine when to sample the data bits.  That leaves only one channel for data, yet there are two (MISO and MOSI, similar to RX and TX).  Not only can you not see both sides at the same time, you also can't see the timing relationship between them.  To me, that's like working in the dark.  And why I was hoping the DS2000 had an undocumented capability (missing from the manual) to use the Ext.Trig for say, Clocking, leaving Chan1 and 2 open for MISO/MOSI.  Not having the 4-th channel (a Select line, for CS or SS) will certainly be detrimental on a multi-slave SPI bus, but that at least you can find ways to work around (have the other slave devices shut-up).  With the DS1000Z, you don't have to.

That's why it seemed like a bit of a stretch to me for Rigol to claim that the DS2000 even handles SPI.  Sure, it can trigger on it (and handle CPOL and CPHA), and decode the bit-stream to turn n-bits into bytes or words... for one half of the comms.  But that's pretty crippled, in my book.

Quote from: echen1024
I am still a bit torn between this scope and the DS1000Z.
I suspect you are far from alone.  They're both great scopes... but each has unique capabilities and limitations.  Thus complicating a decision for those who would much rather have one "best" option.

I've reached out to see if I can get ahold of an evaluation unit, to do a hand's on comparative review of.  I doubt I'd buy one myself, since I already have three 4-channel scopes (and one 8-channel).  But who knows... none are as compact as this unit.  And these are cheap.  Uh, I mean, "surprisingly affordable".  :)
 

Offline iloveelectronics

  • Frequent Contributor
  • **
  • Posts: 940
  • Country: hk
Re: SPI Decoding
« Reply #2 on: November 02, 2013, 05:05:39 am »
I believe it is always in the manufacturer's best interest to NOT create the best got-it-all products, so they keep people wanting more and “upgrade”. Or at least what their sales deaprtments would want them to believe.
My email address: franky @ 99centHobbies . com
My eBay store: http://stores.ebay.com/99centhobbies
 

Offline Mark_O

  • Frequent Contributor
  • **
  • Posts: 939
  • Country: us
Re: SPI Decoding
« Reply #3 on: November 02, 2013, 05:45:19 am »
I believe it is always in the manufacturer's best interest to NOT create the best got-it-all products, so they keep people wanting more and “upgrade”. Or at least what their sales deaprtments would want them to believe.

That's rather cynical, but may have an element of truth to it.  But it's also true that TANSTAAFL applies. 

You can't ever have the best performance, and the best functionality, AND the lowest price.  The DS2072 costs 50% more than the DS1000Z, and has half the channels.  Obviously it's going to win in the performance/features dept, because that's the tradeoff that was made. 

It's pretty amazing to me that the 1000Z is as good as it is, with 4-channels AND at 2/3 the price of the 2000.  Decent 4-channel scopes have never been affordable.  Before now.
 

Offline iceisfun

  • Regular Contributor
  • *
  • Posts: 140
  • Country: us
Re: SPI Decoding
« Reply #4 on: November 02, 2013, 08:43:07 am »
I have had lots of trouble with the Rigol 2xxx SPI decoding stuff, even when my SPI things are working great the scope has trouble with it.

I would have to give the protocol decoder a  :--
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4606
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: SPI Decoding
« Reply #5 on: November 02, 2013, 09:02:25 am »
I found various bugs in the UART, I2C and SPI decoding features on the DS4000 series too. I reported them months ago, but they're not yet all fixed.

Unfortunately Rigol are staying true to the stereotype of Chinese manufacturers in this respect... a decent enough spec sheet, but the finished product just isn't anywhere near as thoroughly tested and reliable as it needs to be. Few things frustrate me more than features that obviously don't work, and which any engineer tasked with testing that feature for 5 minutes couldn't fail to find.

I generally work on the basis that a piece of test equipment needs to be at least 10x as reliable as the equipment it's being used to work on. If an instrument says that a particular condition exists, such as a voltage or a data pattern, I need to be confident that what it's telling me is true - and if it's not completely reliable and trustworthy, then I just have two pieces of kit to debug rather than one.

If you really need serial decoding that works, I'd recommend going with Agilent. The decoding on my 3000X works faultlessly.

Offline Mark_O

  • Frequent Contributor
  • **
  • Posts: 939
  • Country: us
Re: SPI Decoding
« Reply #6 on: November 02, 2013, 10:24:12 am »
Thanks very much Ice and Andy, for the reality behind the claims.  That's very disappointing news.

That's the trouble with evaluating Chinese-sourced scopes from their spec sheets... you can't.  Which is why it's important to actually get your hands on one and try it out.  With the DS2000, and even the vastly more expensive DS4000 having problems, that doesn't bode well for the 1000Z.  :(

Andy wrote:

Quote
...if it's not completely reliable and trustworthy, then I just have two pieces of kit to debug rather than one.

Oh so true.  Sometimes painfully so.
 

Offline Electro Fan

  • Super Contributor
  • ***
  • Posts: 3333
Re: SPI Decoding
« Reply #7 on: November 04, 2013, 03:32:16 am »
I found various bugs in the UART, I2C and SPI decoding features on the DS4000 series too. I reported them months ago, but they're not yet all fixed.

Were some of the bugs fixed?
Any chance you can post the list of bugs you found so we can stay on the lookout for them and the fixes? Thanks
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4606
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: SPI Decoding
« Reply #8 on: November 04, 2013, 07:40:41 am »
Sure. Here's an extract from what I sent back to the distributor a week after taking delivery of a DS4054 with serial decodes enabled:

[snip]

1) MISO and MOSI polarity is not correctly restored on power up, and the UI settings are inconsistent with the levels actually used by the decoder. I have to change the polarity to the opposite sense and back again to get it to decode correctly.

2) SS polarity is wrong. Our SS signal is active low, but the scope only decodes data if I set the polarity to be active high.

3) Bit ordering (LSB / MSB first) is similarly not restored on power up, and must also be set to the opposite setting and back again before it decodes this aspect correctly.

4) The horizontal (time axis) position of the decode becomes inaccurate if the waveform is panned and zoomed. Alignment between the decoded data and the clock that governs its timing is lost.

[snip]

This was all with SPI. The alignment of the decoded data was broken with UART decoding too, as it turned out, which is absolutely no use at all if you need to look at a data stream and see what happened when.

I did get a firmware update which appeared to fix the first three bugs, but not the alignment. Six months later and it's still broken. I don't hold out much hope of it being fixed.

Online H.O

  • Frequent Contributor
  • **
  • Posts: 957
  • Country: se
Re: SPI Decoding
« Reply #9 on: November 04, 2013, 10:58:57 am »
Hi,
I thought the alignment issue WAS fixed in the latest* firmware - at least that was what I found when I (quite briefly) tested the UART decoding on my DS4014 both pre and post that particular FW update a couple of months ago.

* That would be 00.02.00.00.04
 

Offline nowlan

  • Frequent Contributor
  • **
  • Posts: 649
  • Country: au
Re: SPI Decoding
« Reply #10 on: November 04, 2013, 01:39:19 pm »
I was keen for rigol because of protocol decoders, but im thinking zeroplus might be better option. Not sure I need a logicport to snoop comms.
 

Offline Electro Fan

  • Super Contributor
  • ***
  • Posts: 3333
Re: SPI Decoding
« Reply #11 on: November 04, 2013, 07:50:47 pm »
Sure. Here's an extract from what I sent back to the distributor a week after taking delivery of a DS4054 with serial decodes enabled:

[snip]

1) MISO and MOSI polarity is not correctly restored on power up, and the UI settings are inconsistent with the levels actually used by the decoder. I have to change the polarity to the opposite sense and back again to get it to decode correctly.

2) SS polarity is wrong. Our SS signal is active low, but the scope only decodes data if I set the polarity to be active high.

3) Bit ordering (LSB / MSB first) is similarly not restored on power up, and must also be set to the opposite setting and back again before it decodes this aspect correctly.

4) The horizontal (time axis) position of the decode becomes inaccurate if the waveform is panned and zoomed. Alignment between the decoded data and the clock that governs its timing is lost.

[snip]

This was all with SPI. The alignment of the decoded data was broken with UART decoding too, as it turned out, which is absolutely no use at all if you need to look at a data stream and see what happened when.

I did get a firmware update which appeared to fix the first three bugs, but not the alignment. Six months later and it's still broken. I don't hold out much hope of it being fixed.

AndyC_722, Thanks!  If you have any more (about I2C, RS232/UART, or anything else, please post them).

EF
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4606
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: SPI Decoding
« Reply #12 on: November 04, 2013, 07:59:28 pm »
That's pretty much the list of specific bugs I found in the space of a few hours. The rest of my feedback to Rigol was... how should I put this... less technical and more emotive in nature.

Offline echen1024Topic starter

  • Super Contributor
  • ***
  • Posts: 1660
  • Country: us
  • 15 yo Future EE
Re: SPI Decoding
« Reply #13 on: November 04, 2013, 10:15:22 pm »
Looks like it's going to be the 1074Z for me. No money for an Agilent, even though I would LOVE a 4000 series.
I'm not saying we should kill all stupid people. I'm just saying that we should remove all product safety labels and let natural selection do its work.

https://www.youtube.com/user/echen1024
 

Offline Electro Fan

  • Super Contributor
  • ***
  • Posts: 3333
Re: SPI Decoding
« Reply #14 on: November 05, 2013, 12:09:08 am »
That's pretty much the list of specific bugs I found in the space of a few hours. The rest of my feedback to Rigol was... how should I put this... less technical and more emotive in nature.

I can believe that - it takes a good head and a good heart to make good stuff happen; it's a combination I see in your posts and from a fair number of other posters here.  It's part of what makes EEVblog such a deluxe site.   :-+   :-+
 

Offline echen1024Topic starter

  • Super Contributor
  • ***
  • Posts: 1660
  • Country: us
  • 15 yo Future EE
Re: SPI Decoding
« Reply #15 on: November 05, 2013, 12:41:36 am »
Exactly. The quality of the members is superb.
I'm not saying we should kill all stupid people. I'm just saying that we should remove all product safety labels and let natural selection do its work.

https://www.youtube.com/user/echen1024
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4606
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: SPI Decoding
« Reply #16 on: November 05, 2013, 07:24:37 am »
Group hug  :-+  :-/O  :-DMM  :-BROKE


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf

 

- - -
-->