Nemo Excel Don't Download

[magrus]

Hey guys,

I've a Mares Nemo Excel, but when I try to download it shows its downloading but it remains at 0.00% complete.

Tks for your help.

[TnT]

Look here for instructions on how to collect the info we need.

[lerkku]

I´ve had the same problem. It seems that MacDive really does not support Nemo Excel, although it is supposed to. I returned my Nemo Excel to the shop and started to use my old Vyper again.

[TnT]

The libdivecomputer library which is used by macdive for downloading dives, is based on reverse engineering the communication protocol and data format. With a few exceptions we don't get any support from the manufacturers, and therefore there is never any guarantee that it will work flawless. Although for the majority of the devices, especially the ones that are very common and thus well tested, the communication is very stable. Some less common devices like the Nemo Excel, may indeed need some improvement. But since we don't own those devices ourselves, the only way to make that happen is when users are reporting the problem and assist with the data collection.

Now what I need for the Nemo Excel is a capture of the communication between the dive computer and the Mares software. Instructions are available on the libdivecomputer website. Note that you need a 32bit windows (a virtual machine is usually fine too).

[lerkku]

[quote]there is never any guarantee that it will work flawless. Although for the majority of the devices, especially the ones that are very common and thus well tested, the communication is very stable. Some less common devices like the Nemo Excel, may indeed need some improvement/quote]

Here is the problem. If you cannot guarantee compatibility with Nemo Excel, you should not promise that in your specs. Charging money for something that does not work as promised makes customers unhappy.

[nick]

We do our best, and often devices work in some cases and not in others. As Jef says, we get no help from manufacturers so all sorts of things can cause problems.

For example, some devices behave differently when they are full of data - we generally need to actually find someone with a device that is full before we can reliably test this. Or sometimes certain things can be present in a dive that breaks the downloading/parsing of data that we have not seen before. We entirely rely on users to come forward and help out with testing, as it's not feasible for us to buy every dive computer and fill them all up, hit deco on all of them, etc!

Of course, and it should go without saying, I'm more than happy to provide a refund if you are not happy. Email me your order number. There's also a reason for having a trial version - no matter what the software, you should always check you are happy with the trial before purchasing.

At the end of the day, we do expect a little patience. This stuff can be complex and require some effort to get things working. We still find new problems on existing and well tested devices. Without documentation, it's all guess work. But gradually we smooth it out

[TnT]

Here is the problem. If you cannot guarantee compatibility with Nemo Excel, you should not promise that in your specs. Charging money for something that does not work as promised makes customers unhappy.

I can perfectly understand your opinion, but keep in mind that nobody can guarantee an application will never fail. It's a simple fact that ALL non-trivial applications will contain bugs, whether you like that or not.

Regarding the dive computer compatibility, I think you can't deny that the list of supported devices is very extensive. However we don't own every single device on that list, and thus our ability to perform extensive testing is seriously limited. Unless someone is willing to donate a dive computer (because I can't afford to buy them, see below), we have to rely on testing by the owners themselves. It's far from ideal, but it's usually our only option.

The way this works is that if someone would like to see his/her dive computer supported, we provide instructions on how to collect all the data we need. Next, we try to figure out how the download protocol works, and build a first prototype application that the user can try. Finally, after a (time consuming) process of making changes and further testing, we end up with something that works for the test person. At this point we add it to the list of supported devices and make it available for everyone to try. No doubt there may be new problems discovered. But the only way to find out is when users like you do report them. And that's only possible when you can actually try it out. It's a chicken and egg problem.

I'm sure most people here will prefer a reasonable functional downloader with an occasional bug here and there, rather than having no support at all.

BTW, you may not be aware of the fact that the libdivecomputer library, of which I happen to be the main developer, is an open source project that is available to everyone free of charge. So technically you did pay for macdive, but not the libdivecomputer library. These are two separate projects, although we do cooperate very closely.

[lerkku]

I do understand the difficulties you developers are experiencing. Nevertheless, if you cannot be certain about a dive computer´s compatibility, you must state it clearly in the specs (e.g. "sometimes compatible" or occasionally supported), doing otherwise is misleading. Take me for instance, if I had known that download from Nemo Excel does not work properly , I wouldn´t have bought the instrument in the first place.

[TnT]

I agree that it would be great to know the compatibility level for each device, but unfortunately we can't provide an accurate indication. The reason is that we don't know in advance whether a problem will ever be discovered or not.

Take for example your Nemo Excel. Before this topic was posted, there were no known problems with the Nemo Excel, so it would have been listed as fully supported. The end result would have been exactly the same as now.

I don't even have numbers on how much testing each device receives. People usually only contact me to report problems, not when everything is working well. But when I get many such reports for a particular device, that could equally well mean that it's a very popular device. Take for example the group of the Oceanic Atom2 compatible devices. Downloading works very well, but because each variant introduces subtle changes in the dataformat, I get many reports of small errors in the parsed output. Thus this backend receives by far the most bugfixes, although it is very stable and used with success by many people.