Page 1 of 1

Quad - Thoughts and Observations

PostPosted: Mon Jan 31, 2011 7:11 am
by BenF
Finally I had some time to actually test the Quad and here’s a brief summary of my thoughts and observations (this is an engineering pre-production sample). A handheld DSO is somewhat novel as there aren’t many competitors around with specifications and pricing matching the Quad. So doing a comparison is difficult, but one appropriate question to ask might be if this is a toy for the hobbyist or a tool for the professional?

Hardware

The four probe connectors are smaller than on the Nano and probably appropriately so as the unit packs a lot of features in a small package. Overall design looks and feels good. I like the top four push buttons, but less so the two slide left/right buttons. They are in my opinion too small, feel flimsy and prone to fail. They may tolerate more “abuse” than I think, but this is for time to tell. The oscilloscope style probe looks well made, but the black and red hook probes again are too small and delicate for normal use (they may however be appropriate for SMT probing).

The quad has two analog channels (72Msa/s) and two logic only channels (sample rate unknown). Both for hobbyist and professional this is a useful combination to have in a single instrument. User interface requirements however differ significantly between analog and logic channels so it will be quite a challenge to get this right for both. Sample rate is sufficient for a wide range of practical measurement challenges (analog waveforms up to about 7MHz).

One might expect that primary use for a handheld DSO is in the field. I doubt however that the Quad will face up to the elements well and so may be more appropriate for the lab. Further to this, I find the storage capacity of 2MB too small (not expandable as far as I know). When in the field you want to record measurements, bring along sample waveforms and then you’re likely to wish for more. For lab use this is less of an issue since you’re likely to have access to a laptop/desktop where you can upload measurements.

I’m missing a real time clock that would allow me to time stamp saved waveforms and measurements.

My sample came without a power switch (the slider is missing).

Software

Hardware defines the overall feature set and the precision/accuracy we can achieve in terms of measurements. This is typically the main (and quite often the only) aspect considered prior to buying. Software on the other hand defines the users experience and to a large extent determines what customers will think of the product once they actually own it (will they use it and will they recommend it to others).

Sadly to say, the current firmware is in a horrible state. It appears to be based on the Nano 2.5e version with all the mistakes of the past present in abundance. This looks like something a hardware engineer would put together in a couple hours just to test as much of the functionality as possible from a single screen. Navigating the various options take you on a wild tour around the full display with no apparent logic. Knowing where to go for what is just about impossible to guess and if you have an idea, getting there is even worse.

The only feature I could get to work with some success was displaying a repeating waveform with AUTO trigger. At T/Div’s less equal to 1us however, the waveform was lost and the display started to live a life of its own. Nearly all other options are either not working at all or doing something different than what I would expect.

At this stage I think further testing is pointless. First priority should be to get the basics right. That is sampling, buffer management, triggers (AUTO, NORM and SING) and waveform display. Until this is right, nothing else is relevant. In my experience it is always better to include a small feature set that works well rather than a large feature set with lots of issues.

Conclusions

Going back to the initial question (a hobbyist toy or a professional tool) I’m leaning towards the hobbyist and student. The hardware has plenty of features that would cover the needs of professionals and hobbyists alike, but I’m afraid exposing it to the elements is more than it can handle.

As I see it, the present state of the firmware prevents any practical use of the Quad as a DSO and measuring instrument. Get the basics right, remove all non-working features, seek help with the user interface (think in terms of menu hierarchies and use cases) and then let us test again.

Re: Quad - Thoughts and Observations

PostPosted: Tue Feb 01, 2011 4:23 am
by dnordenberg
Hi

I feel exactly like you about the quad, mainly because of the software and how you operate it. I didn't know about the push button functions on the left/right switches until justblair detected it, not very user friendly and probably a big reason why they feel "flimsy".

Wonder why seeed started from nano firmware 2.5e then? I suspected for a long time that seeed does not have any control over the development, especially not the software. Everything is done by someone else outside seeed (bure e-Design Co.,Ltd.?) and seeed just have to accept what they get and market that. e-Design seem to run the minidso site which does not seem to have any connections to seeed? e-Design may not even know about the BenF firmware. Why is there always hardware design differences between e-Design and seeed? Did seeed develop it and e-Design made a copy or is it the other way around? In my posts in the past I have tried to gather facts of the relationship between seeed and e-Design but any such post is unanswered by seeed. And we have never seen a real DSO software or hardware developer from seeed here on the forum (AFAIK). Only people from seeed that have been able to make smaller hacks in the firmware, clearly not the original developers.

Don't want to sound rude or back talk seeed but something is strange here. It can't be seeed who is developing the nano or quad. Nothing strange with that, this is the point with open source hardware or software, anyone can do what they want with it but I get the impression that seeed is trying to make us think that they are the ones making everything on their own or is it me that got everything totally wrong here?

Re: Quad - Thoughts and Observations

PostPosted: Wed Feb 02, 2011 4:40 am
by justblair
Seeed have always credited others for the designs of the nano and quad, the designer is Chai Xiao Guang. Just as Seeed carry designs by Dangerous Prototypes and others, I don't think Seeed are being underhand. I agree that they could be more forthcoming on the details of collaborations they are involved with. At the same time, I have posted links to other sites like miniDSO and no attempt has been made to censor my posts... so I dont think they are trying to obfuscate the truth.

The miniDSO site would appear to be closer to the original developer (It appears that this is the original source of the firmware apparently), but the fact that Seeed had engineering samples at all would indicate that they have more than a passing association. I notice also that the miniDSO site appear to have missed out the V2 Nano in their range.

Ben... What version of the software have you used? Not that there is a huge difference between the versions, but 0118 seems the most reliable to me so far...

I think that if the software is heavily based on the Nano would that mean that new improved software might be easier to develop? I hope that you have plans to dabble.

2ps worth

PS. Thanks for your comment on the other thread Ben, I was really worried that I might have offended, kind of you to put my mind to rest.

Re: Quad - Thoughts and Observations

PostPosted: Wed Feb 02, 2011 11:26 am
by dnordenberg
Chai Xiao Guang is bure. So the e-Design DSOs are in fact the originals? Is bure the developer of both the hardware and the software? It would seem so as Seeed writes "Designed by Chai Xiao Guang" and the source is full of e-Design associations.
Maybe I went a bit to far in previous post, it was not meant as accusations, just pure thoughts. I also think it may be just that they are too fuzzy with the collaboration details. Seeed always talk in "we" form, not "e-Design" or "e-Design and we" so it is very unclear who did what. Most recent example seeed writes "but we over-clocked AD9288-40". With dangerous prototypes they are more clear with texts like "we cooperated with some other firms like Dangerous Prototypes". Seeed simply never talk about e-Design for some reason.

Re: Quad - Thoughts and Observations

PostPosted: Mon Feb 07, 2011 1:52 pm
by ESP
Sorry guys, we are just back from Chinese New Year and away from forums...
Please first excuse us for sending out buggy samples while striving to bring it to beta version. Especially on the software part, it's still in alpha stage. We are focusing on the hardware part and SYS code now, trying to make it as much stable and efficient as possible. The APP layer carrying UI will be open sourced as DSO Nano soon, it's just a simple stack up of functionalities.

@BenF:

Thanks for the sharing and suggestions!
We are trying to find a balance of size/cost/convenience. 72Msa/s analog channel requires serious probes, while the digital channel requires less. We don't want to include 5 big probes which is a mess to carry. We might need to customize for smaller and neat probes but it takes quite some time on sourcing. (the best we found is mueller, now shutting down...)

We are not confident to call it a professional tool at all, it's born in shade of more powerful and cheaper ICs, and evolving with open source community. It's our goal to bring it up step by step to offer lessor professional tool for everybody, in long term.

@dnordenberg

Sorry for not stating this clearly before, we were a bit worried about the grey market impact. We never meant to obfuscate the truth, let me clarify here.

As in the package you have noticed, DSO series are developed by Chai Xiaoguang (Bure) and his team (e-design), he is a very veteran Chinese Engineer. Seeed Studio's role is more like a producer of a movie, we coordinate the resources, bring it up to reality products, collect the feedback and work with e-design to improve. It's similar to other cooperative projects like with dangerous prototypes.

But it's not simple as working with dangerous prototypes. Since Seeed Studio is mainly covering oversea Markets, e-design are distributing kits to China hobbyist markets. Due to the price difference, someone assembled the kits and resell via e-bay, as the DSO201 you may have seen. To better distinguish them and upcoming domestic OEMs (like http://www.china-victor.com/cn/product_ ... 196&ID=229 ), we changed the enclosure and package with a few improvements. MiniDSO site is the domestic(China) channel recently built by e-Design.

In the future, Seeed Studio will be the sole portal to oversea marketing. You will see "Innovated by e-design" in the laser marks, and Seeed Studio is working with e-design on a series of DSO with more open hardware infrastructure.

Please let me know if this answers the doubts, thank you!

Re: Quad - Thoughts and Observations

PostPosted: Mon Feb 07, 2011 2:06 pm
by tomwid
Thank you, that was very helpful information!

Re: Quad - Thoughts and Observations

PostPosted: Sun Feb 13, 2011 9:45 am
by lygra
BenF wrote:The only feature I could get to work with some success was displaying a repeating waveform with AUTO trigger. At T/Div’s less equal to 1us however, the waveform was lost and the display started to live a life of its own. Nearly all other options are either not working at all or doing something different than what I would expect.
When I tried to use the Quad today I found similar issues. The Quad capture buffer software is pretty much usless in it's current form.