Is Rigol DS1054Z still a good option in 2019?

[Martin72]

And then there has to be someone else who finds out, thank you for your commitment.

[artur0089]

10-12-14-16 bits are a deception of marketers for gullible consumers. Very similar to Mega-Giga-Tera pixels in camera matrices.
3% vertical channel error, are they serious?!! DS2072A - 2%, development for more than twelve years!!!
A good Logic Analyzer replaces any decoders in an oscilloscope.
The noise in my DS2072A is the same as in the 12-bit DHO8-9 from Rigol.
DHO-HDO does not "search" for anomalies in large amounts of memory.
DHO8-9 no Hi-Res! DS1000Z 12 bit Hi-Res...
DHO8-9 small font.

[Martin72]

What are you trying to tell us...

10-12-14-16 bits are a deception of marketers for gullible consumers.

Proposal:
You feed in a small signal, let's say a sine with 100mVpp, then you press stop and resolve to vertical and post the result.
A DHO800 owner does the same and also posts the result.

3% vertical channel error, are they serious?!! DS2072A - 2%, development for more than twelve years!!!

I think you meant the gain accuracy...

DC Gain Accuracy[4]
± 1% (>5mV/div, FullScale)
± 2% (≤5mV/div, FullScale, Typ.)

Or did you mean the offset...

DC Offset Accuracy ≤200 mV/div (±0.1 div ± 2 mV ± 1.5% of offset value)
>200 mV/div (±0.1 div ± 2 mV ± 1.0% of offset value)

The noise in my DS2072A is the same as in the 12-bit DHO8-9 from Rigol.

By the way, what is he like?

DHO8-9 no Hi-Res! DS1000Z 12 bit Hi-Res...

LOL...DHO8-9 IS 12 bit Hi-Res...

That was probably nothing, whatever you wanted to say.

[Fungus]

DS1000Z 12 bit Hi-Res...

At what bandwidth...?

[Smokey]

DS1054Z is OK for bench use without remote apps. All DS1000s have a USB bug that makes mass data transfer impossible.
The new DHOs are nice, however, so maybe that one is better.

I used to dump full captures over ethernet.  Went fast enough with no issues.

[artur0089]

Proposal:
You feed in a small signal, let's say a sine with 100mVpp, then you press stop and resolve to vertical and post the result.
A DHO800 owner does the same and also posts the result.

Another marketing ploy. No one ever “stretches” a saved (stopped) oscillogram vertically! “Stretching” horizontally is normal, this is why you need more memory.

I think you meant the gain accuracy...

Vertical Accuracy:
TDS700D - 1% - 8 bit.
DS2000A - 2% - 8 bit.
DS1000Z - 3% - 8 bit.
DHO & HDO - 3% - 12 bit.
It's like making a DMM 6½ characters, but leaving the accuracy at 4½...

By the way, what is he like?

Noise comparisons must be made in the same room and on devices with the same frequency band! My device is hacked to 300 MHz.
Limit 100 MHz.


There is no noise in an open field, deep forest or in the middle of the ocean...

DS1000Z 12 bit Hi-Res...

At what bandwidth...?

≥5 μs/div, up to 500 kHz this is enough.

DHO8-9 no Hi-Res! DS1000Z 12 bit Hi-Res...

LOL...DHO8-9 IS 12 bit Hi-Res...
That was probably nothing, whatever you wanted to say.

The HDO1000 and HDO4000 have Hi-Res 14 and 16 bits.
DHO800 and DHO900 do not have this!
--------------------------------------
LOL...
The conversation with you is over...

[memed]

Displaying four curves in parallel, let alone 16 logic traces on the DHO900,  is really cramming the usable screen area; this was one of the reasons I excluded these models from consideration.

What model did you choose then?

There is something like a horizontal time setting, you could use that, every scope has it.

This was my theory but wasn't sure if it work same in practice without any hassle and deformation of wave.

Maybe better to have more detail horizontally.

Whatever: It's obviously the way things are going so you might as well get used to it.

Yes you are right, checked many other models and look like this is new standard now.
So I'm considering now DHO904 or wait for bargain on DHO1074
Unless find another model to confuse me

[ebastler]

Displaying four curves in parallel, let alone 16 logic traces on the DHO900,  is really cramming the usable screen area; this was one of the reasons I excluded these models from consideration.

What model did you choose then?

I bought a DHO1074 during the Black Friday sale: Larger screen overall, and a larger proportion of the vertical space is actually available for trace display. I was really pleased with the display, but ended up returning the scope due to other issues. Waiting for the SDS1000X HD now...

[memed]

How much was black friday deal for this model(DHO1074) that time? What problem you had with that oscilloscope?

[Veteran68]

How much was black friday deal for this model(DHO1074) that time? What problem you had with that oscilloscope?

Not speaking for ebastler, but it was crazy cheap. Like $599 US and I believe the same in Euro. With my eevBlog Saelig discount I think I paid $562 US with shipping but before sales tax. I've since upgraded it to an $1849 US equivalent scope. Crazy deal for <$600.

Yes, it's buggy. There are threads here full of the issues identified. I'm sure I'd be happier with the SDS1000X HD, and depending on its price when released, I may sell my DHO1074 to buy it.

That said, it does indeed have a beautiful screen, and honestly the bugs haven't (yet) hindered my use. But I'm a hobbyist, and not a sophisticated user. So YMMV.

[ebastler]

How much was black friday deal for this model(DHO1074) that time? What problem you had with that oscilloscope?

The Black Friday discount on the DHO1074 was very tempting -- 40% off list price, which meant 720 Euro incl. our 19% VAT. I realize that I will have to pay quite a bit more for the SDS1000X HD when it eventually arrives in Europe.

I did not find any showstopper bugs, but increasingly got the impression that Rigol just doesn't care. Some examples:

• The FFT is just as rudimentary as in my 9-year-old DS1054Z (no Averaging & Peak mode), and has the same bugs (messed-up window functions causing distortions of the peaks).
• Setting parameters via the dual Flexknobs was less useful than I thought: They operate at odd increments and without acceleration, so I found myself using the on-screen keyboard much more often than I liked. The knobs also got confused occasionally as to which parameter they were controlling, indicating a horizontal adjustment but changing vertical scale instead etc.
• Acquisition rates in Recording mode were painfully slow; less than one shot every 5 ms. And the indicated time stamps did not seem real, they did not match stop-watch time.
• Trigger coupling in AC or low-frequency rejection mode is useless, it has a massive jitter. Looks like nobody at Rigol ever tried this.
• The memory upgrade key which was included in the deal took three reminders to Rigol before it arrived, and then did not work at all. (Even though I had the proper instructions, courtesy of the hacking thread. With the non-documentation which Rigol includes, there's no chance in hell you could make even a functional key work.)

Combine that with the fact that I had doubts regarding the longer-term future of this model: The Black Friday deal felt like a closeout sale, and the scope has not received the same firmware improvements which the DHO800/900 got. I sent it back within Amazon's generous "no questions asked" return window.

If you are considering the DHO800/900 series, I would strongly encourage you to wait for the SDS800X HD from Siglent. It's a much more capable scope, although with a slightly lower-resolution screen. Not shipping in Europe yet, but Performa01 (who is a beta tester) has just started to publish an in-depth review series: https://www.eevblog.com/forum/testgear/sds800x-hd-review-demonstration-thread/

[Martin72]

Hi,

It's a much more capable scope, although with a slightly lower-resolution screen.

Resulution is the same (DHO800 1024x600, SDS800 1024x600).

[ebastler]

Resulution is the same (DHO800 1024x600, SDS800 1024x600).

Oops, I was thinking of the DHO1000 with its higher resolution screen (while the Siglent counterpart, SDS1000X HD, also has 1024 * 600). You are right, the competing smaller models have the same screen size and resolution.

Text on the DHO800 nevertheless looks smoother than on the SDS800X HD since it is rendered using anti-aliasing. But the control and display areas also take up more room and leave less for the trace display.

[Fungus]

The FFT is just as rudimentary as in my 9-year-old DS1054Z

But it's much more detailed, has much faster update rate, much lower noise floor, it's much easier to move the trace around and rescale it (touch screen!), and (most importantly) it has a table showing frequencies and amplitudes of the peaks.

Acquisition rates in Recording mode were painfully slow; less than one shot every 5 ms.

That's going to depend totally on your configuration and memory size. There are slow cases and fast cases, just like on every other oscilloscope.

Trigger coupling in AC or low-frequency rejection mode is useless, it has a massive jitter. Looks like nobody at Rigol ever tried this.

Can't say I've ever used that in 10 years of oscilloscope usage (I had to look up what it did when it was reported) and, given how long it took for it to be reported, I suspect hardly anybody else uses it either, including the people at Rigol.

I've always triggered from what's visible on screen because it's visual and you can see what you're doing.

IOW it's almost a non-bug in reality.

That said, it does indeed have a beautiful screen

Yep.

and honestly the bugs haven't (yet) hindered my use. But I'm a hobbyist, and not a sophisticated user.

Exactly.

If you're a "sophisticated user" then get a Siglent SDS2000. This 'scope isn't aimed at you.

(and neither is the SDS800 when it finally arrives)

[Martin72]

(and neither is the SDS800 when it finally arrives)

It will arrive and relatively soon.
And it will be a direct competitor in terms of price, at least at the 70Mhz level, but not in terms of performance, where it is superior to the Rigol, as you can see in Performa01's new thread.
I didn't expect this clear superiority.

[ebastler]

Acquisition rates in Recording mode were painfully slow; less than one shot every 5 ms.

That's going to depend totally on your configuration and memory size. There are slow cases and fast cases, just like on every other oscilloscope.

Believe it or not, I realize that. So when I came across the slow recording mode, I tried it with very favorable conditions: just channel 1 enabled, 2 µs/div, 1 kpts memory, simple edge trigger. Recording 1000 frames, the average time step between stored records was 5.8 ms according to the scope's timestamps.

And even those timestamps are apparently wrong: Acquiring a series of 1000 frames took 14 seconds on a stopwatch, while the progress bar and frame counter moved forward in a linear manner. But the scope claimed that the last record has a timestamp of 5.8 seconds?!

I am not ruling out a user error, since the recording is so absurdly slow. But egonotto observed the same "performance", we both published our findings here, and nobody reported different results. Does the DHO800 fare better?

[thinkfat]

[...]

3% error doesn't mean that the higher vertical resolution is useless. Accuracy is not noise. If the noise was drowning out the 8 lower bits - yes, that'd be problematic.

You still get a more truthful waveform reproduction, less vertical aliasing noise and that is interesting for signal analysis. OK, your measurements have 3% maximum error. But a DSO is not about measurement accuracy anyway. There's a reason why some DSOs have built-in multimeters.