FNIRSI DLA-16 Plus / DLA-32 Plus Logic Analyzer

[megabyte]

I am an FNIRSI DLA-16 Plus owner and would like to share some info about it.

Product page:	https://www.fnirsi.com/products/dla-32
User manual:	https://cdn.shopify.com/s/files/1/0694/8310/2426/files/DLA-16_Plus_DLA-32_Plus_Logic_Analyzer_User_Manual_EN-V1.2-2026.03.17.pdf?v=1786704068

Specs:

Feature \ Model	FNIRSI DLA-16 Plus	FNIRSI DLA-32 Plus
Number of Channels	16ch	32ch
Max Sampling Rate (Buffer)	1CH/1G, 8CH/1G, 16CH/500M	1CH/1G, 8CH/1G, 16CH/500M, 32CH/250M
Max Sampling Rate (Stream)	1CH/1G, 2CH/1G, 4CH/500M, 8CH/250M, 16CH/125M	1CH/1G, 2CH/1G, 4CH/500M, 8CH/250M, 16CH/125M, 32CH/50M
Min Collectable Pulse Width	2ns	2ns
Max Collectable Frequency	200MHz	200MHz
Hardware Storage Depth	4Gb	4Gb
Max Sampling Rate (Buffer Stream)	4Gbit At low sampling rates, depends on the computer's memory size	4Gbit At low sampling rates, depends on the computer's memory size
Data Cache Mode	Buffer / Stream / Roll	Buffer / Stream / Roll
Acquisition Accuracy	±1 sampling cycle	±1 sampling cycle
Trigger Type	Rising Edge / High Level / Falling Edge / Low Level / Any Edge	Rising Edge / High Level / Falling Edge / Low Level / Any Edge
Protocol Support	170+ (open protocol decoding library, customizable for any protocol)	170+ (open protocol decoding library, customizable for any protocol)
Input Withstand Voltage	-40V~+40V	-40V~+40V
Input Impedance	250K/15pF	250K/15pF
Threshold Range	-5V ~ +5V (0.1V step)	-5V ~ +5V (0.1V step)
Waveform Output	3.3V, 1Hz~20MHz, 4-channel PWM	3.3V, 1Hz~20MHz, 4-channel PWM
Channel Interface	1.27mm shielded cable interface	1.27mm shielded cable interface
Power Supply Voltage	5V±0.2V	5V±0.2V
Power Supply Current	460mA (@5V max)	460mA (@5V max)
Dimensions	≈90*81*11.5mm	≈90*81*11.5mm
Operating Temperature	0~70°C	0~70°C

PS: Teardown photos coming soon.

[0xdeadbeef]

Hm, Software is "based" on Sigrok, but it doesn't look like they released the sources. Which is most probably violating Sigrok's license.
At least with DSView for the DSLogic Devices, all the sources and it's directly supported by Sigrok anyway.
I would stay away from this as long as there is no official support inside Sigrok or the sources for the software are released.

[megabyte]

Front



Back

[Doctorandus_P]

Hm, Software is "based" on Sigrok, but it doesn't look like they released the sources. Which is most probably violating Sigrok's license.

fnirsi lies and cheats in everything they make, and they intentionally hide hardware deficiencies (such as a scope not being able to work at some frequencies (which are within the "claimed" bandwidth b.t.w.)) with averaging and other software fuzzing. So I don't expect them to treat the Sigrok / Pulseview project in any different way.

I really like the Sigrok / Pulseview project very much, and I find it very sad to see that very little development is happening with it. I would gladly pay 10% extra for logic analyzer hardware if that 10% goes into supporting the Sigrok / Pulseview project.

[0xdeadbeef]

At least the design is no direct clone of the DSLogic, but OK, Pango is a Chinese FPGA brand, so they save quite a bit there, I guess.
Anyway, I guess the hardware part could be OK or even good value and a Sigrok based software is better than some loveless Chinese crap SW, but still...

BTW: are there any advanced trigger options? Can't find anything in the manual.
And does the internal capture allow RLE compression like on the DSLogic?
if neither is the case, the value of the hardware is somewhat crippled by the (closed source) software.

And the manual says you can use negative thresholds down to -5V. Is that really true? Is there any usecase for this?

[megabyte]

@0xdeadbeef and @Doctorandus_P, I totally agree with both of you. Moreover there is no Linux software for this LA.

I took Wireshark+USBPcap and tinkered a Sigrok driver for the LA.

[BBBbbb]

Hm, Software is "based" on Sigrok, but it doesn't look like they released the sources. Which is most probably violating Sigrok's license.
At least with DSView for the DSLogic Devices, all the sources and it's directly supported by Sigrok anyway.
I would stay away from this as long as there is no official support inside Sigrok or the sources for the software are released.

DS did that only after they were forced and are just dumping blobs, they are not better than fnirsi.

[0xdeadbeef]

From what I can tell, the source code for DSView and even for the FPGA are released, but admittedly I never looked at it to see if this is really complete or usable:
https://github.com/DreamSourceLab
But at least the current situation is much better than with a closed source Fnirsi SW and no direct Sigrok support.

[BBBbbb]

From what I can tell, the source code for DSView and even for the FPGA are released, but admittedly I never looked at it to see if this is really complete or usable:
https://github.com/DreamSourceLab
But at least the current situation is much better than with a closed source Fnirsi SW and no direct Sigrok support.

It’s better because of Sigrok, not DS.
They were forced to comply with the license. Tried the same as fnirsi

They are no better!
If they really wanted to be better they’d be contributing to pulseview and not just being a parasite.

[megabyte]

are there any advanced trigger options?

The Windows software from FNIRSI has a tab for sophisticated triggers, but it is disabled. I haven't delved into this topic yet. In the moment the standard capabilities of Sigrok are more than enough for me.

And does the internal capture allow RLE compression like on the DSLogic?

I do not know. The streaming mode is enough for me.

And the manual says you can use negative thresholds down to -5V. Is that really true?

I am no sure 100%, but probably yes it is true. The LA accepts a 16-bit value as threshold voltage, so in theory it can be negative and be in compliance with the specs (+/- 5V with 0.1V step).

Is there any usecase for this?

I have no idea.

[voltsandjolts]

Front
(Attachment Link)
Back
(Attachment Link)

Chinese Pango FPGA
Chinese "Cypress EzUSB" clone (probably)
Chinese PUYA FLASH
Chinese 3PEAK OpAmps
...
Impressive how the Chinese are developing their own independent semiconductor industry (OK, with a bit of RE assistance, aka copying).
I wish Europe could be organised enough to do the same.

Oh - Google tells me there are 120 fabs in Europe, which is much more than I thought. And more than the 80ish that China has..but they may have more throughput and more modern processes.

[megabyte]

Chinese "Cypress EzUSB" clone (probably)

I suspect that the role of USB 3.0 bus controller plays WCH CH569.

[voltsandjolts]

Ah, yes it could be.
I was think of the classic FX2 clones, but this FNIRSI device is USB3. And there is JTAG port.

Are you planning to share your sigrok driver?

[megabyte]

Are you planning to share your sigrok driver?

Yes, for sure. But first, I need to clean up the code.

[Sorama]

Are you planning to share your sigrok driver?

Yes, for sure. But first, I need to clean up the code.

Can you share how you bought it?
From the Fnirsi site/shop?

They do charge more than 20€ of taxes, right?

[megabyte]

Can you share how you bought it?
From the Fnirsi site/shop?

They do charge more than 20€ of taxes, right?

1. I ordered the LA directly on fnirsi.com. FNIRSI states that they cover taxes and customs clearance costs for goods priced up to $150USD for some EU countries. See the paragraph "International Shipping" on https://www.fnirsi.com/pages/shipping-policy. Germany is in the list.
2. I entered the discount code published in the order page.
3. I did not include any add-on like "Worry-Free Purchase".
4. I paid €148,82 EUR in total via Paypal. It is the final price. German VAT (MwSt) was included in the total price. I did not pay any additional taxes or/and customs clearance costs in Germany.
5. The delivery was to my private address.
6. The delivery was from China to Germany and took approximately 2 weeks.

[Doctorandus_P]

Impressive how the Chinese are developing their own independent semiconductor industry (OK, with a bit of RE assistance, aka copying).

yes, the chinese are working very hard to become as much independent of "the west" as they can.

A long time ago I almost bought a DS-Logic, but I did not buy it because of the atrocious way they treat the Sigrok / Pulseview project. So I managed with the little EUR 10 boxes, as most of what I do is UART / SPI / and such at moderate baudrates and 8 channels is often enough too.

A few years ago I bought a Kingst LA2016 because it is Pulseview compatible (They do not mention Sigrok / Pulseview at all, but have their own independent software), but Pulseview has more decoders.

Recently I bought a spare LA2016 and both the FPGA and the uC had been replaced with local chinese versions.

For logic levels, there are some protocols that work with negative voltages. The good old RS-232 goes to -12V (although uart with standard logic levels are more common these days). Another example is ARINC 429, which uses positive and negative pulses.
https://en.wikipedia.org/wiki/ARINC_429

It is however not common connect such signals directly to a logic analyzer. It's much more common to use a hardware level converter, such as an MAX232, RS485 or CAN transceiver, etc on a small adapter board to convert the signals before they go to the LA.

[TomKatt]

So - roughly 50% price of the DSLogic U3Pro16...

I have the older DSLogic Plus (usb 2.0) version and while it does make me somewhat uneasy how they took advantage of Sigrok, for a long time it was one of the best bang-for-the-buck LA's.  At least they eventually opened up their code fork.  Though I understand they haven't really contributed back to the original project and managing two forks is very complicated to the point of almost being no help at all to Sigrok.

I'm also a musician and these products remind me of Behringer lol.  I am ashamed to own them, but sometimes you need to do what you need to do.  I will say the DSLogic Plus has worked very well for me.

PS - This review aligns with my concern about the FNIRSI software...  The hardware is one thing, but many times the software makes or breaks the overall experience...  I'm not sure what their history of addressing software issues is like, but I wouldn't hold my breath for substantial improvements.

Great hardware, software still needs improvement

I really like the DLA-16 Plus. The device looks excellent, feels solid and well built, and the hardware has a lot of potential.
However, the software still needs some refinement. I found a few bugs, some decoder stability issues, and some usability details that could be improved. I was also disappointed to discover that the Advanced Trigger function was incorrectly listed in the product specifications.
FNIRSI support has been responsive and acknowledged the specification error, which I appreciate.
Overall, I think this is a very good logic analyzer with strong hardware, but the software needs a few more improvements to make the product truly excellent.

[voltsandjolts]

Yeh, the screenshots I've seen of the FNIRSI software aren't great, but maybe it works well enough.
They could have saved themselves the effort and instead just published a driver for sigrok pulseview.

I too have the Kingst LA2016 and 1016.
The pulseview driver was very broken when I purchased (or Kingst had changed the protocol) so I contributed driver fixes and full schematic.
That prompted a maintainer (gerhard) to buy a LA5016/5032 and implement support - very similar to LA2016, in fact same driver is used.
https://github.com/sigrokproject/libsigrok/tree/master/src/hardware/kingst-la2016

[TomKatt]

Yeh, the screenshots I've seen of the FNIRSI software aren't great, but maybe it works well enough.
They could have saved themselves the effort and instead just published a driver for sigrok pulseview.

It seems like a no-brainer that would have been the thing to do - everyone benefits and they don't have to write and maintain their own proprietary software.  Especially as I suspect their software 'borrowed' from Sigrok behind the scenes anyway.  I wonder if a deep code analysis was done what would be found...

[0xdeadbeef]

Especially as I suspect their software 'borrowed' from Sigrok behind the scenes anyway.  I wonder if a deep code analysis was done what would be found...

There no need to "suspect". They clearly state that the software is based on Sigrok on their product page. Which is why a commented about this some days ago.

As a side note: for all the LAs I owned which also featured Sigrok/PulseView support, the Sigrok driver never contained all the features of the original software.
E.g. for the Ikalogic ScanaPLus, only the most basic capture and trigger features were working and - as far as I recall - the RLE feature for the DSLogic is also still missing when using PulseView instead of DSView.
Besides, Sigrok development was pretty much dead in the last years anyway. When I posted a link to a repository with a working SENT decoder with lots of examples in the mailing list two years ago, nobody even cared to reply.
It's still reassuring to have a Sigrok driver as a backup solution of course, but unless the driver is feature complete, it's really not much more than that.

[megabyte]

As promised, I'm publishing a Sigrok driver for this LA.

Just having AI "tinker" the driver wasn't enough — it contained bugs. I spent considerable time testing it as thoroughly as I could, using several dev boards and peripherals across different buses. All bugs found have been fixed.

The driver is https://github.com/megabyte1024/libsigrok.

Summary:

• The driver was written by AI.
• It was tested by me, with AI's help, on several dev boards and different peripherals.
• The driver is partly designed to support both the "DLA-16 Plus" and "DLA-32 Plus", but currently only works with the "FNIRSI DLA-16 Plus".

[TomKatt]

As promised, I'm publishing a Sigrok driver for this LA.

Just having AI "tinker" the driver wasn't enough — it contained bugs. I spent considerable time testing it as thoroughly as I could, using several dev boards and peripherals across different buses. All bugs found have been fixed.

Thanks for your work!

Just curious - were you able to determine if the FNIRSI LA's implement RLE compression for data efficiency?  I tried to look that up but I don't see any references to RLE.  Though as they seem to be pretty much copying existing products (which incorporate RLE), I'd guess they'd also copy that?

[megabyte]

Just curious - were you able to determine if the FNIRSI LA's implement RLE compression for data efficiency?

I am sorry. At the moment, I do not plan to delve into this.

I'd guess they'd also copy that?

It’s pure speculation, but it probably that this logic analyzer is based on or inspired by the logic analyzer DL32 from Alientek.

Facts are:
- Similar LAs specs,
- Software for both LA looks similar.
- Both software use similar protocols.

[TomKatt]

It’s pure speculation, but it probably that this logic analyzer is based on or inspired by the logic analyzer DL32 from Alientek.

I had not seen that model before...  yes, it seems more than a bit identical ;-)

Both are a bit vague on some specs, but the FNIRSI page lacks any mention of RLE encoding while the Alientek page infers RLE encoding in their comparison chart (though they may simply be stating the obvious that RLE is more efficient, not that they implement it lol):


EDIT - I never understand Aliexpress pricing because it seems to change depending what device I'm using to access the net and what pages I've been looking at...  I swear they apply some kind of real-time pricing strategy..  BUT it looks like the Alientek DL32 version is selling for $325 which is a good $65 more than the FNIRSI model