I may as well give my views, since I supported the Kickstart project.
Resolution:
What is often overlooked, or possibly ignored, is that more resolution on a device than what it is primarily designed for, is not a benefit. The resolution picked for this unit was, IMHO, an intelligent and well thought out choice. More resolution means more work for the device, meaning likely decreased performance where the resolution isn't required, and which could also impact battery life. Artifacting from downscaling, considering that it will, for a large percentage of the small segment where it is necessary, basically be displayed in "half" resolution. The screen isn't large, so this should not be offensive. While some performance could be impacted by the downscaling, these titles would probably still be able to enjoy a performance boost if the emulation is written to discard half the information, rather than process it and not display it. For odd resolutions, like EGA, this might not be as pretty, but it's a pretty small segment to be extremely concerned about.
Accuracy:
I have to say that, while I understand and appreciate the concerns of folks looking for true accuracy, I think it's a bit misplaced when applied to a device like this one. On a full size rig, I too want to see the best possible representation. But this is an OTG device, and one with a fairly small screen. Fully accurate MAME functionality would be great, but it's not a necessity. One could just play ports to the lower console systems, but they never really stack up to the originals. In the end, even with the limitations in earlier versions of MAME, the experience will still be more satisfying than what the ports can deliver, and as long as the hardware can do that well, it will have succeeded in the most important way.
What makes this more interesting than the usual fare:
This is simple. At least the appearance of an exuberant developer base, and a group dedicated to furthering the cause. This is something which, as mentioned before, is very hit or miss with the typical Chinese devices of this nature. The Dingoo has enjoyed a decent amount of support, and it looks like those who developed for it liked it enough to go to work on the hardware side to get past issues they might not ordinarily be able to get past. This is a "good thing", as it indicates a commitment to the platform. It also looks like they tried to gauge the hardware base to the task at hand, so that developers for emulation wouldn't be left wanting in the ways which are the most important. I have a GP2x, which was also supposed to be an emulation powerhouse, and it wasn't bad for it's time. But even with the support it had, it still wasn't as good as many had hoped. Community support was short-lived, due to new hardware obsoleting the older, and the poor choice of directional controller doomed it once the better ones came along. So now it lives on my personal "Island of Mis-fit Toys". The PSP was the next stop, and it's a great device. But the sketchy nature of modifying the firmware severely limits the audience for both developers and users. The system is proprietary, and what is known about the internals had to be hacked to be discovered, or gleaned from an official developer sympathetic to the cause. This is another hindrance which causes development for the platform to go stale, and one which isn't present with the GCW. Android, on the other hand, seems very promising. It's open, and there's no shortage of inexpensive devices which run it and run it well. But it too has some concerns. The fact that the hardware bases are so plentiful and varied makes it hard to get the most out of any particular device being developed for, without risking compatibility with others. And with the Chinese manufacturers cranking out a half dozen new ones every week, this problem isn't getting any better. So while one could theoretically just pick one to focus on, and push the community in that direction, without financial commitment to the device there is no guarantee that the device will continue to be available, making all of that investment of time, short on return.
In a sense, the folks with the GCW are doing what I just described as the key for Android, and are going about it in the right way. A standardized, and fairly powerful, hardware base which they can continue to have manufactured so long as there is demand, and as they seem committed to the device, they have the tools to help fuel that demand. If they continue to squash the bugs in the emulation, and deliver a solid performing unit, they will cause a "snowball" effect on the size of the user base, which will in turn, help to guarantee continued development. I believe something not so different is why MAME is where it is today. If there was no interest (demand) there would not have been so many talented and dedicated individuals involved in the project. So my feeling is, if the concept of the GCW is one which appeals to you, and you were already considering a Dingoo (which I was), or any other handheld for emulation purposes, why not give it a chance?