Chinese company “Ainol” have released Novo 7 — sub-$100 7″ tablet with Ice Cream Sandwich (Android 4.0) and MIPS processor. This is quite big step aside from previous ARM-powered models, and the price speaks for itself. MIPS and ARM are different beasts, but Ainol have already got Google Compatibility Certificate and endorsement by Andy Rubin, so it looks like most Android software runs just fine on the new tablet.
Screen size is standard 800×480, capacitive touch screen, HDMI output and everything else is just as expected.
I’ve already ordered one on ebay for ¥11,320 JPY (including delivery), and hope to get it before holidays.
Two other major delivery services are supported now by mail tracker: Sagawa Takkyubin (佐川宅急便) and Kuroneko (ヤマト運輸), but it’s still as simple as it was before — just enter the number and the program will automatically check and decide where and how to track it. Barcode scanner works now almost always, success rate is about 95% even in extreme lighting conditions. It takes only 2-5 seconds to scan a receipt and get all the tracking info available on the screen. If you send a lot of mail you know how troublesome and time consuming this might be.
Here’s version 0.9.7 (still beta, but improving quite fast and steady): 配送追跡.
Got some spare time and made a little application for Android, which helps to check the current status of my mail and packages right from my phone. Actually I’ve tried a few other similar applications first, but haven’t found the one which works just as I like. Most of them don’t support barcode scanning and don’t validate input, so I could input any garbage instead of tracking number (you know how error prone is touch-typing =) and then spend hours wondering why my parcel does not show up in the tracking list. Another thing I don’t like is to choose every time what kind of number I have, Expack or cash on delivery/registered mail — they all look very similar written on the paper.
My application (配送追跡) takes the number and selects the proper category automatically, also it calculates checksum for every number and rejects anything that fails validation. Took me a few hours to reverse engineer checksum calculation for Japan Post tracking numbers, but finally it works, and, by some strange coincidence, same validation scheme is used for Sagawa (佐川宅急便) and Kuroneko (ヤマト運輸) express delivery tracking numbers, so maybe I’ll add support for those in the future.
There are still a few kinks to iron out, and current barcode scanning library from “ZXing team” gave me mixed results on my phone, but people say it’s the best one and the actual problem is in my phone, so I hope it works well for you.
And, please, don’t worry about the kanji on the screenshot — the program will display everything in English if you phone has non-japanese locale set.
For those worried, it’s not like I’m jumping the ship, but more like grabbing another opportunity and expanding to the new devices. After I’ve paid $99 to the “empire of evil”, I’ve got a nice formatted e-mail, saying: “To complete your purchase and access your Apple Developer Program benefits, please click on the activation code below.”
Clicked on the link and there’s another message, telling me they have some problem and I should contact support. Support sent me another message, which says “We have reviewed your documentation and have removed the hold on your Program activation. At this time, we ask that you please refer to the original activation code email that you received and click through the link once again.”
Well, “If at fist you don’t succeed, try, try again!”. Plenty of fun and games and for only $99.
After a long while, I’ve recently started to code a few things for Android, launched Eclipse and was greeted with mysterious errors. Which include:
- “layoutlib is too recent. update your tool” in the Eclipse layout editor. This was solved by upgrading android libs and SDK and also upgrading ADT from inside of Eclipse->Help->Upgrade. The former upgrade probably was not really necessary, but I did it anyway just in case to prevent other incompatibilities from coming out.
- another gem appeared when I tried to create an empty project to test a few ideas: “error generating final archive: Debug certificate expired”. This required deleting ~/.android/debug.keystore and “cleaning” the project, which eventually led to regeneration of required debug keys. Google says debug keys are expired after 365 days, so it must have been more than 1 year since I’ve started playing with Android.
- if you plan to access internet from the emulator, open Run->Configuration->Target and set “Additional Emulator Command Line Options” to “-dns-server 18.104.22.168,22.214.171.124″ and don’t forget to require INTERNET permission in the manifest.
And yes, I’ve got two walking plastic Androids to race them against each other around my table.
I’ve got a few questions about the geiger graph plots and how to make them. Here’s a short explanation for those without extensive computer background. Personally I prefer using RRDTOOLS to plot any data graphs, but the installation under Windows might be a bit too difficult for the average windows user. Anyway, if you plan to user RRDTOOLS, here’s the download link, and don’t forget you’ll need to install CYGWIN as well. Japanese users might appreciate someone already had translated the manual.
First, you have to launch your favourite terminal application and get some data from the (USB-)serial port and copy it with (Ctrl/C) or any other available technique. This is the data you’re going to make a graph from.
Open Excel or any other spreadsheet software and paste the data into it. Don’t forget to set “separators” to spaces or the data won’t get separated into the different columns.
In the column just before the data, enter two consecutive time values (13:30 and 13:31, for example), select them and drag black square with the mouse to fill the whole column with the time values.
This is new, totally reworked board which can accomodate any tube, from SI-3BG through SI-37/39G upto SI-1G and even SBM-20, though the last one is a bit too large and leaves tips hanging off the board. The tube on the picture is SI-39G. High voltage generator is now controlled from the microprocessor, instead of fixed setting as on previous boards and also there’s a noisy buzzer, which ‘ticks’ just like the original geiger counter.
New software has changed a lot, now includes voltage controller, buzzer support and fast screen update. The data over USB is still sent once a minute, but the LCD display is now updated every 4 seconds (2-3 seconds were also possible, but seemed to be overkill) with the data of the last minute. Instead of fixed 1 minute intervals, this version uses “sliding 1 minute window” to calculate the value. There’s no need anymore to wait for full 1 minute to see readings go up after introduction of some radiation source, the change is obvious and imminent. Though, on the down side, this fast pace could be a bit disconcerting to some users.
Recently I’ve got a few SI-1G tubes, so naturally I’ve built a new board to accomodate this tube, which is about three times as big as SI-3BG. Well, the schematics is almost the same, capacitors went up a grade and basically only the board layout had changed a little.
This tube has almost everything I need for a household radiation measurement post — high sensitivity, low required voltage and very good response even to the smallest amounts of radiation.
I have brought home a piece of granite (which has about 0.2uSv/h natural background radiation level) and put it next to the SI-1G around 13PM on Monday. As I said before, granite does emit mostly alpha and tube has been rated for gamma only, so I did not expect much.
But, almost immediately I have noticed a +4CPM spike which is still continuing since the rock stays near the tube until now. I have not seen this good response even from the Chinese-made dosimeter which costs several times as much. I’m off to bed and wonder if the high level of secondary gamma emission will wear off around tube by tomorrow morning (I hope not =)
ps. Ah, and thorium mantle makes it really scream at whopping 255-280CPM, which converts to something in 2-3uSv/h range — scary!
It took quite a while to find the relatively safe radiation source. Thorium mantle from camping supply store is usually sold and used for camping gas lamps, but does emit some alpha radiation. Alpha is relatively safe, because it can hardly penetrate a sheet of paper or a plastic, so as long as I don’t eat or inhale it I should be fine.
Actually, the usual dosimeters cannot detect alpha, but can detect secondary gamma and beta emissions which happens when alpha strikes something. Therefore, I did not expect any precise measurement, but it was quite reassuring to get the numbers pretty close to the cheap Chinese dosimeter I’ve borrowed from a friend of mine.
Well, last week has shown the total lack of any radiation around my home. I have got pretty stable reading for a few days, which moved a little bit higher and lower, but mostly consisted of random background-level noise. The noise (~0.2 clicks per second), which should be expected from this kind of geiger tube at their respective age.
While this kind of picture looks quite dull and boring, I’m kind of cowardly reluctant of ordering any kind of radioactive materials to my home, so I’ve decided to take another approach. I’ve heard the granite stones are supposed to have 0.3-0.5uSv/hour levels, so probably tomorrow would be a good time to visit a stone sculpture factory nearby.
And for that purpose, here comes another board, this time without any USB connection, but equipped with a small LCD screen, so I can attach a pair of batteries and use it to see the results while away from home. Also I’ve got a cheap (relatively) Chinese dosimeter, so there’s a hope I can calibrate the output of my boards and don’t rely anymore on thumb-sucking approach when calculating CPM-to-uSv/h conversion rates.