Over four years ago I wrote about my weather station setup. Well ... It doesn't exist anymore. There was a severe thunderstorm and my gadgets weren't properly fastened to the steel pipe and this happened:

After weather cleared, I realized the lack of readings from my garage roof. The poor temperature unit / CPU was upside down, hanging from some wirings and FULL OF WATER! Ufff! Electronics and water won't mix. At least not well.

My zip-ties didn't like the ping pong -ball sized hail:

The picture from sunroofing is from my friend's place. He has some tinkering to do, to get this deck roofing fixed. That piece of plastic didn't survive chunks of ice plummeting from the sky.

To get my weather properly measured and data-logged, I already put a purchase order for a new unit. This time I'm going for a Davis Vantage Vue. Also, to save money this time I'm attaching it properly using steel hose clamps!

It's not that you'd be talking all the time, but quite often a good microphone is needed. It may be (video)conferencing to somebody or just to get your multiplayer game coordinated with the other people you're playing with. Lot of people are happy using the tiny mic in their headphones, but not me. I'd rather invest some real money to a proper piece of hardware to get my sound transmitted over crystal clear.

Since I didn't want to go for ridiculously expensive studio ones, I stayed below 300 € price range and went for a Røde NT-USB (details are at https://www.rode.com/microphones/nt-usb). The microphone would work as a self-standing desk mic, but adding a few options allows me to amp up the experience to a near-studio -quality.

This is what my rig looks like:

As you can see, there is a PSA1 boom arm and a SMR shock mount to keep the sound quality better when I bang the desk in rage.

Mic itself looks like this:

The metal ring used to fasten the microphone to a stand gives you indication of the quality of this product. No messing around here. No flimsy plastic screws. Steel!

Also the acronym USB gives you a hint about how you might connect this lump of metal to a computer, using USB 2.0 interface on the bottom of the microphone. If this would be professional-quality, instead of USB-B connector, you would use for example XLR-connector for cabling. This would leave you to do the analog-to-digital -conversion by yourself. I chose this particular unit for the reason it hooks up to your PC/Mac/Linux/whatever directly. Here no extra stuff needed, but the setup is limited by USB's 5 meter max. cable length. There is plenty of USB-cable in the box, so don't worry about running out.

Shock mount is for reducing any possible noise caused by accidental headbump into desk or microphone. As suggested, the mount absorbs most of the shock and not transmitting really weird noises over. This is how the "cradle" looks like:

There is a windshield reducing any air blowing from me towards the microphone. Again, reducing any weird noises sounding like I was hit by a hurricane to being transmitted over. The shock mount connects to the boom arm using a 5/8" (or 3/8") screw:

As you can see, there is an adapter there allowing you to use pretty much any mic boom of your choice. I chose to go full on with Røde.

Important thing is to note, that manufacturer does NOT recommend using SMR shock-mount with NT-USB microphone. Product details at https://www.rode.com/accessories/smr do NOT list NT-USB as being compatible or related product. The reason is not as dramatical as you might think. The products are compatible, but:

In a scenario, where you would use the USB to transmit sound back to your headphones via 3.5mm jack in the mic, it would be difficult for you to change the headphones volume as the volume knobs aren't that easy to access. You can access them, but not as well as manufacturer would love you to. So, they simply flag the products as "incompatible".

Ultimately, I've been really happy with the product. My sound is heard extremely well on the other side. Helping with this is the boom arm, which lets me place the microphone to almost any location of my choosing while doing my computing, whatever type of computing that might be. The best thing is the very high quality of the product. I don't think I'll be buying a new mic anytime soon.

Randomness in computers

You don't need to know much about computers to understand, that computers cannot do random things. Yes, all programming languages and libraries do offer you a rand()-function to emulate randomness. However, the resulting output will follow the carefully crafted programming implementing this "randomness". The most trivial pseudo-random functions will merely provide a sequence of numbers appearing random, but this sequence can be reset to start from beginning making the "randomness" predicatable. That's not really very random, huh!

Improved randomness in computers

To be fair, there does exist improved pseudo-random algorithms which take their initial seed-values from something volatile (time is one such volatile parameter) making the quality of randomness better. Still, even high-quality pseudo-random algorithm is just complex sequence of operations, which will produce duplicate results on same input values. Sometimes its just very tricky to craft a situation where all of the input values would match.

If somebody is capable of doing that, your randomness changes into predictability. Read the story of Dual_EC_DRBG on Wikipedia https://en.wikipedia.org/wiki/Dual_EC_DRBG. When you're generating your precious private keys, you don't want anybody (including NSA) to be able to guess what you have there.

Random source in Linux

Since a proper random source is something every single user, developer and sysadmin would love to have, the problem has been approached on your Linux by authors of the operating system. An excellent description can be found from Wikipedia article https://en.wikipedia.org/wiki//dev/random#Linux. Briefly put, your Linux will collect environmental entropy from number of sources (including human interaction with keyboard and mouse) to a pool, which can then be used to produce naturally random numbers. It actually works very well, the quality of randomness is top-notch.

Obvious problem with this approach is, that you cannot pull too many random numbers out of this source without exhausting it. The fix is to keep typing something while moving your mouse (not a joke!) to generate entropy for the random source. This will eventually help fill the entropy pool and /dev/random will spit couple bytes more.

Those users who have exhausted their /dev/random on an idling rack server without a console keyboard, mouse and video know that it takes painfully long for the entropy pool to fill. A busy server doing something will be able to fill the pool much faster.

A real random source

If you need a real proper random source, which works without human intervention and can provide really good randomness as a stream, there are possibilities on hardware. I know of two good ones, Simtec Electronics Entropy Key and ubld.it TrueRNG Hardware Random Number Generator.

Note: if you consider getting one, get the TrueRNG version 3 (http://ubld.it/truerng_v3). Its just that I have the 1st gen version at hand and haven't found the reason to upgrade.

My TrueRNG looks like this:

It is essentially an USB-stick.

Linux lsusb info essentially identifies it as a Microchip (vendor ID 0x04d8) manufactured USB-device (with ID 0xf5fe) providing RS-232 communications:

Bus 002 Device 009: ID 04d8:f5fe Microchip Technology, Inc.
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.00
  bDeviceClass            2 Communications
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0         8
  idVendor           0x04d8 Microchip Technology, Inc.
  idProduct          0xf5fe
  bcdDevice            1.00
  iManufacturer           1 ubld.it
  iProduct                2 TrueRNG
  iSerial                 0
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength       0x0043
    bNumInterfaces          2
    bConfigurationValue     1
    iConfiguration          0
    bmAttributes         0x80
      (Bus Powered)
    MaxPower              100mA
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           1
      bInterfaceClass         2 Communications
      bInterfaceSubClass      2 Abstract (modem)
      bInterfaceProtocol      1 AT-commands (v.25ter)
      iInterface              0
      CDC Header:
        bcdCDC               1.10
      CDC ACM:
        bmCapabilities       0x02
          line coding and serial state
      CDC Union:
        bMasterInterface        0
        bSlaveInterface         1
      CDC Call Management:
        bmCapabilities       0x00
        bDataInterface          1
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x81  EP 1 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x000a  1x 10 bytes
        bInterval               1
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass        10 CDC Data
      bInterfaceSubClass      0
      bInterfaceProtocol      0
      iInterface              0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x02  EP 2 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x82  EP 2 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               0
Device Status:     0x0001
  Self Powered

And by looking at /dev/, there is a /dev/ttyACM0. That's how udevd will populate a CDC-device when it sees one.

How is this a "true" random source?

Oh, that's easy. The device will produce a random 0 or 1 bit constantly when its on. Or to be precise, there is an internal algorithm producing those based on a constant flow of electrons on a transistor PN-surface. The exact phenomenon is called avalance effect or avalance breakdown. For those who can do electronics, there is a good explanation about this in Difference Between Avalanche Breakdown and Zener Breakdown (I borrowed the visualisation pic from above link).

To (over)simplify that, in a carefully constructed electronic circuit, inside a transistor an electron may or may not be emitted on the other side of a semiconducting surface. The occurrence is as random as it can be in nature. Other circuitry will detect this random flow of electrons (or lack of flow) to produce ones and zeros.

What makes this a really good for randomness, as it is well established that this avalance of electrons will happen. Also, it will happen often enough to produce a stream of events. It's just that we don't know exactly WHEN the avalance of electrons will happen. If you time-slice this to slots, a slot can be empty (no avalance) or full (electrons avalanching).

Linux tweaking:

udev

Anybody having multiple devices in their Linuxes knows, that you really cannot control which device name some specific device will get on reboot. To overcome that, udevd can be instructed to do things when it sees a device. My rules for TrueRNG include setting it to highest possible speed and creating a symlink-device so, that I can point to a known source of random. Also, I'm loosening access to that source of randomness to any users belonging to dialout-group. If I wouldn't do that, only root would have access to this fine random-source.

My /etc/udev/rules.d/99-TrueRNG.rules contains:

SUBSYSTEM=="tty", ATTRS{product}=="TrueRNG", SYMLINK+="TrueRNG", RUN+="/bin/stty raw -echo -ixoff -F /dev/%k speed 3000000"
ATTRS{idVendor}=="04d8", ATTRS{idProduct}=="f5fe", ENV{ID_MM_DEVICE_IGNORE}="1", GROUP="dialout", MODE="0664"

If you want to take your random-device for a spin, you can do something like:

dd if=/dev/TrueRNG of=random.bytes bs=64 count=1024

That would create a file of 64 KiB containing very very random bytes. In theory you can just cp data out of the character device, but since it has an infite flow, you'll need to cut it at one point.

rngd

Remember the part I said earlier about Linux using your keypresses and mouse movements as entropy source for randomness. Even with the USB-stick popped into a PC, that still remains the case. What needs to be done next is to offer a helping hand to the Linux kernel and make sure the entropy pool is always full.

My Fedora has package called rng-tools. It is packaged from Mr. Horman's https://github.com/nhorman/rng-tools. What's in there are the tools for pumping those precious truly random bits out of the USB-source to Linux kernel's entropy pool. As default, rngd will use /dev/hwrng as the source for randomness. Some Linuxes don't have that device at all, some Linuxes point that into CPU's random source. What's guaranteed, it will not point to your USB-stick! We need to change that.

Btw. you might be horrified by the fact, that something is fidding with your randomness. The exact bits transferred from USB to entropy pool won't be the actual bits getting out of /dev/random. Your keypresses and many other events are still a factor. Its still a good idea to not run randomness-monitoring malware or spyware in your Linux.

Systemd works so, that I did create a copy of /usr/lib/systemd/system/rngd.service into /etc/systemd/system/rngd.service. The contents of the copy in /etc/systemd/system/ can be freely modified and it has priority over the /usr/lib/systemd/system/ one. The only change I made was to have the ExecStart-line say as:

ExecStart=/sbin/rngd -f --rng-device=/dev/TrueRNG --fill-watermark=4000

When rngd-service would be started, it will use the USB-stick as source and make sure, there are at least 4000 bits of entropy in the pool.

Making sure rngd setup works

At any given point, you can query how many bits are available in the Linux entropy-pool:

cat /proc/sys/kernel/random/entropy_avail

Since my setup is working correctly, it will display a number greater than 4000 and smaller than 4096. The upper limit comes from /proc/sys/kernel/random/poolsize, which is a hard-coded number from Linux kernel source.

Hint: If you do the stupid thing like I did and set the /proc/sys/kernel/random/write_wakeup_threshold (using --fill-watermark) into 4096 (or above), your rngd will keep hogging CPU like there is no tomorrow. It is impossible for the pool to contain maximum number of bits at any given time. Give your system a break and set the threshold bit lower than max.

Finally

It's always nice to know for a fact, that random numbers are random. This fact can be verified and has been verified by number of other people.

Enjoy!

Every one of use will eventually end up in a situation where old mobile device is upgraded to a new one. Most of us can manage getting the new one working without problems. There are backups somewhere in the cloud and the new device will have the intial setup done from the backup and everything in the out-of-the-box experience will go smoothly.

But what about the old one? You might want to sell or give or donate the perfectly functioning piece of hardware to somebody who wants it more. But not without your data, accounts and passwords!

Apple support has this HT201351 article What to do before you sell, give away, or trade in your iPhone, iPad, or iPod touch. It instructs you to do a full erase of the device, but doesn't go too much into the details.

Personally, I've struggled with this one a number of times. So, I decided to record the full sequence for me and anybody else needing it. Here goes!

Start with Settings, General. In the bottom of General-menu, there is Reset. That's your choice. For curious ones, you can go see the menus and choices without anything dangerous happening. You WILL get plenty of warning before all your precious data is gone.

In Reset, there are number of levels of reseting to choose from. You want to go all the way. To erase every single bit of your personal data from the device. To get that, go for Erase All Content and Settings.

You will have two separate warnings about your intent to destroy your data. Even if you pass that gateway, there is more. Nothing will be erased until a final blow.

The final thing to do is to inform Apple, that this device won't be associated to your Apple ID anymore. For that, your password will be needed. This is the final call. When you successfully punch in your password,  then the big ball starts rolling.

When you see this, you're past point-of-no-return.

It takes a while to erase all data. Wait patiently.

When all the erasing is done, the device will restart and it will go for the out-of-the-box dialog. This is where new user needs to choose the user interface language, network and associate the device with their own Apple ID.

A while back I got my hands on a nice PCIe 2.0 network interface card.

The thing looks like this:

In theory, that's an extremely cool piece to run in one of my Linuxes. Two 10 gig ethernet ports! Whoa!

Emphasis on phrase "in theory". In practice it's just a piece of junk. That's not because it wouldn't work. It does kinda work. I have to scale down the DDR3 RAM speed into 800 MHz just to make the hardware boot. Doing that will get 12 GiB RAM out of 16 available. Something there eats my PCIe lanes and forces them to work at unacceptable low speeds.

This is a serious piece of hardware, for example in Amazon.com the going price for such item is $340 USD. Given the non-functional state of this, I got this with a fraction of it. Given my personal interest in such toys, I had to go for it. This time it didn't pan out.

Maybe HPE support site phrase for this at https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-c03663929 explains it all: "PCIe 2.0 compliant form factor designed for HPE ProLiant servers". I'm not running it on a HP ProLiant.

On a Linux, it does work ok. Kernel driver bnx2x detects and runs the NIC instantly. Linux lspci info:

# lspci -s 02:00.0 -vv -n
02:00.0 0200: 14e4:168e (rev 10)
  Subsystem: 103c:18d3
  Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
  Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- SERR-   Latency: 0, Cache Line Size: 64 bytes
  Interrupt: pin A routed to IRQ 17
  Region 0: Memory at f4000000 (64-bit, prefetchable) [size=8M]
  Region 2: Memory at f3800000 (64-bit, prefetchable) [size=8M]
  Region 4: Memory at f4810000 (64-bit, prefetchable) [size=64K]
  Expansion ROM at f7580000 [disabled] [size=512K]
  Capabilities: [48] Power Management version 3
    Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold-)
    Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=1 PME-
  Capabilities: [50] Vital Product Data
    Product Name: HPE Ethernet 10Gb 2P 530T Adptr
    Read-only fields:
      [PN] Part number: 656594-001
      [EC] Engineering changes: A-5727
      [MN] Manufacture ID: 103C
      [V0] Vendor specific: 12W PCIeGen2
      [V1] Vendor specific: 7.15.16
      [V3] Vendor specific: 7.14.38
      [V5] Vendor specific: 0A
      [V6] Vendor specific: 7.14.10
      [V7] Vendor specific: 530T
      [V2] Vendor specific: 5748
      [V4] Vendor specific: D06726B36C98
      [SN] Serial number: MY12---456
      [RV] Reserved: checksum good, 197 byte(s) reserved
    End
  Capabilities: [a0] MSI-X: Enable+ Count=32 Masked-
    Vector table: BAR=4 offset=00000000
    PBA: BAR=4 offset=00001000
  Capabilities: [ac] Express (v2) Endpoint, MSI 00
    DevCap: MaxPayload 512 bytes, PhantFunc 0, Latency L0s <4us, L1 <64us
        ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset+ SlotPowerLimit 75.000W
    DevCtl: CorrErr+ NonFatalErr+ FatalErr+ UnsupReq+
        RlxdOrd- ExtTag+ PhantFunc- AuxPwr+ NoSnoop+ FLReset-
        MaxPayload 128 bytes, MaxReadReq 512 bytes
    DevSta: CorrErr+ NonFatalErr- FatalErr- UnsupReq+ AuxPwr- TransPend-
    LnkCap: Port #0, Speed 5GT/s, Width x8, ASPM L0s L1, Exit Latency L0s <1us, L1 <2us
        ClockPM+ Surprise- LLActRep- BwNot- ASPMOptComp+
    LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- CommClk+
        ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
    LnkSta: Speed 5GT/s (ok), Width x8 (ok)
        TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
    DevCap2: Completion Timeout: Range ABCD, TimeoutDis+, LTR-, OBFF Not Supported
         AtomicOpsCap: 32bit- 64bit- 128bitCAS-
    DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis-, LTR-, OBFF Disabled
         AtomicOpsCtl: ReqEn-
    LnkCtl2: Target Link Speed: 2.5GT/s, EnterCompliance- SpeedDis-
         Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS-
         Compliance De-emphasis: -6dB
    LnkSta2: Current De-emphasis Level: -3.5dB, EqualizationComplete-, EqualizationPhase1-
         EqualizationPhase2-, EqualizationPhase3-, LinkEqualizationRequest-
  Capabilities: [100 v1] Advanced Error Reporting
    UESta: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
    UEMsk: DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
    UESvrt: DLP+ SDES+ TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol-
    CESta: RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr+
    CEMsk: RxErr- BadTLP+ BadDLLP+ Rollover+ Timeout+ AdvNonFatalErr+
    AERCap: First Error Pointer: 00, ECRCGenCap+ ECRCGenEn- ECRCChkCap+ ECRCChkEn-
        MultHdrRecCap- MultHdrRecEn- TLPPfxPres- HdrLogCap-
    HeaderLog: 00000000 00000000 00000000 00000000
  Capabilities: [13c v1] Device Serial Number d0-67-26---------------
  Capabilities: [150 v1] Power Budgeting
  Capabilities: [160 v1] Virtual Channel
    Caps: LPEVC=0 RefClk=100ns PATEntryBits=1
    Arb: Fixed- WRR32- WRR64- WRR128-
    Ctrl: ArbSelect=Fixed
    Status: InProgress-
    VC0: Caps: PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
        Arb: Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
        Ctrl: Enable+ ID=0 ArbSelect=Fixed TC/VC=01
        Status: NegoPending- InProgress-
  Capabilities: [1b8 v1] Alternative Routing-ID Interpretation (ARI)
    ARICap: MFVC- ACS-, Next Function: 1
    ARICtl: MFVC- ACS-, Function Group: 0
  Capabilities: [1c0 v1] Single Root I/O Virtualization (SR-IOV)
    IOVCap: Migration-, Interrupt Message Number: 000
    IOVCtl: Enable- Migration- Interrupt- MSE- ARIHierarchy-
    IOVSta: Migration-
    Initial VFs: 16, Total VFs: 16, Number of VFs: 0, Function Dependency Link: 00
    VF offset: 8, stride: 1, Device ID: 16af
    Supported Page Size: 000005ff, System Page Size: 00000001
    Region 0: Memory at 00000000f4820000 (64-bit, prefetchable)
    Region 4: Memory at 00000000f48a0000 (64-bit, prefetchable)
    VF Migration: offset: 00000000, BIR: 0
  Capabilities: [220 v1] Resizable BAR
  Capabilities: [300 v1] Secondary PCI Express
  Kernel driver in use: bnx2x
  Kernel modules: bnx2x

If anybody has any suggestions/ideas what to try, I'll be happy to test any of them. Also, my suggestion for anybody planning to get one is to NOT pay any money for it.

Most of us have a credit card (or debit) for payment purposes. As I love tech, technically speaking, it is laminated polyvinyl chloride acetate (PVCA) piece of plastic conforming to ISO/IEC 7810 ID-1.
What's in a card is:

• The standard-compliant plastic frame
• ISO/IEC 7816 smart card with 8-pin connector
• ISO/IEC 14443 RFID chip and antenna for contactless payments
• (optional, deprected as insecure) ISO/IEC 7811, 7812 and 7813 compliant magnetic strip

The reason I got interested about this begun when I got a new debit card. Anybody having one of these cards know, that they do expire eventually. My bank sends me new one roughly one month before expiry, at which point I tend to destroy the old one into very small pieces to avoid some garbage digging person to be able to exploit my information.

This time, I took couple of pictures of it first (then I destroyed it) to publish in the Internet.
Waitaminute!
You're not supposed to post an image of your credit card! See https://cheezburger.com/8193250816 for a my-new-credit-card fail.

No, I'm not going to do that, instead:

Notice how couple years of usage made the card crack and laminated back and top parts of the card started peeling off. The lamination process failed somehow. Maybe the superglue wasn't super enough, or something similar. I've never seen such a thing happen before.

Here goes: I publicly posted pictures of my old debit card! Obviously, before doing that, I redacted my card number. Also, I'm not going to publish image of the back side having the CVC validation number and my signature. For those curious why I din't redact all 16 digits, the first four are not that important, because its kinda obvious the card is a Visa (first digit 4) and in Finland all Visa cards are issued by Nets Oy (formerly Luottokunta). For card numbers, see https://stevemorse.org/ssn/List_of_Bank_Identification_Numbers.html, a page rejected by Wikipedia, but resurrected from Archive.org

Points of interest in card images:

1. ISO/IEC 7816 smart card
   • All of the chip in the top slice of the card.
   • The chip is bit thicker then the top slice of the card. There is a shallow dent in the back slice to make room for the smart card chip.
2. RFID antenna for contactless payment.
   • It's EVERYWHERE! I never realized how much antenna is required to power the RFID chip.
   • If you would follow the silver antenna, it would make a very long track around the card. It will never cross. This is required to form a long loop. For those not familiar with physics and electricity, it forms a solenoid which will produce current when moving in a magnetic field (payment reader).
   • There is antenna both sides of the back-slice. Sides are connected in two points, 2 and 3.
   • Most of the antenna is around the 7816 smart card chip. That's why people are instructed to put the smart card into contactless payment terminal.
   • I think (please correct me, if I think wrong) the RFID-chip is very near the smart card chip.
3. I think (please correct me, if I think wrong) there are ten capacitors to first absorb electric current from the payment terminal magnetic field via solenoid and store it into the capacitors for the chip to do it's magic of EMV-payment. The transaction will last less then a second, so not much is needed.
4. The back-side antenna
   • Connected to front-side antenna in 2.

Bonus

If you really, really want to, you can tear your card apart and make a ring out of it:

Read all about that Man dissolves credit card to make contactless ring.

This is where an angle grinder meets aluminium.

Angle grinding through a 3.5" hard drive case, spinning platter and controller electronic circuitry will produce enough heat and metal scraps to produce an absolutely unrecoverable hard drive. The magnetic coating containing the actual data doesn't like any kind of dirt, metal chips nor rapid heating.

Funny thing: One of the IBM drives was from era of glass platters. I had no idea about that at the time. Ear protection and noise made sure of that. Afterwards the glass platter was shattered into small pieces making the hard drive sound like a jingle bell. There was a small time slot when glass was used in drives to improve data density. Later manufacturers learned to smooth the aluminum platters beyond glass ones and abandoned the brittle materials.

Electronic equipment recycling - expect me to show up with couple pieces of dismantled electronics.

Humble Bundle was giving away GRID 2 for free: https://www.humblebundle.com/store/grid-2. Price was right and me, as a greedy bastard, of course leeched the ware.

GRID 2 is an arcade racer. It has some level of simulation, but mostly it is targeted for regular gamers. Still, driving a racing car with a typewriter doesn't really sound sensible to me and I whipped up my G29 wheel and started the game.

Crap!

Back in 2013 they didn't support my wheel. Turns out both Codemaster's forum ([Logitech G29] Can't use wheel with GRID2 & GRID Autosport) and Steam forum ([Guide] Add G29 and G920 support to Grid Autosport (Includes Clutch and shifter)) have lengty discussion about tinkering the wheel to work. However, given the age of the game, the XML-file referenced in discussion is long gone. The information is valid.

First: This is an obvious one. Make sure you have Logitech Gaming Software installed. Danger: The wheel seems to install correcly in Windows 10, but it won't. You really really need to have the Logitech stuff installed also. I got version 9.02 from https://support.logitech.com/en_us/product/g29-driving-force/downloads.

Second: The XML-file everybody is talking about is here: lg_g27.xml. Place it into steamapps\common\grid 2\input\ (suggest backing up the original file first!) and then selecting Direct Input Device from game controller options. This enables choosing of Preset <lg_g29> and further tailoring controller settings to your own liking.

Happy racing!

Lately, I've been playing Mudrunner: Spintires a lot. That's a really weird game about logging machinery and mud. For some strange reason I find hauling logs from a lake through mud really entertaining. Why logs are located a "lake" is beyond me.

During the gameplay, player needs to operate number of different heavy machinery. This involves driving a skidder, a heavy truck or operating a crane to get the logs loaded. I tried doing that with a traditional mouse and keyboard config. I kinda managed to do it, but anybody who has driven a real car or truck or flown an airplane using only a keyboard knows, it's very tricky. Then I got a hint, that a Playstation 4 controller is really good with that game.

A-haa! All I need to do is hook up my PS4-controller into my PC using ... well... dunno. As Google search engine is a good friend, I found a project DS4Windows from Github https://github.com/Ryochan7/DS4Windows. It will convert a PS4-controller into an Xbox controller, which will work perfectly with most games.

Just hokk up an USB-cable to the controller, run the software, and:

You're done! It really is that simple. Now spinning tires in mud is much much more enjoyable!

The other day I realized, that my mobile subscription has a voicemail. Since its a corporate one, I really cannot control it much, it just ... well ... is.

This existence of voicemail became evident for a simple reason. Somebody left me a message there. Ufff! Why! Personally I hate those 70s relics. If you cannot get a hold of me, just send a SMS or an email! ... People... They just don't get it.

So. I had an easy task, go listen the message. In an iPhone, there is a Voicemail menu:

Tapping the "Call Voicemail" does absolutely nothing. Eventually I did find the voicemail number from my subscription paperworks. That is, if I really was interested in the message, but I wasn't. Being a geek, I wanted to get my iPhone to do what the button caption says. To Call Voicemail.

Apparently that is something my Telco should fix. Or, that's how its done in the Big World. No such joy for me. After a good while of googling around, I found bunch of codes, which actually didn't either work, or do anything useful on an iPhone. Finally, I bumped into an ancient Apple community discussion with topic Q: iphone 5 change/add voicemail number? This user had been instructed the following:

I have just been in contact with Orange regarding this they advised to dial:
*5005*86*123#

This will set the voicemail number to 123 (orange) so hopefully (i've not tested) you can just put your voicemail number inplace of 123!

Yes! Doing a *5005*86*<my voicemail number here># does the trick for my iPhone 8 running iOS 12.1.

While at it, I googled more and found article iPhone codes from the same era.

For a complete list of generic mobile phone Unstructured Supplementary Service Data (USSD), or "Quick Codes" or "Feature codes" see Wikipedia article https://en.wikipedia.org/wiki/Unstructured_Supplementary_Service_Data#Code_table.

Now that my new PC is done (see my previoius post and Larpdog's Twitch https://www.twitch.tv/videos/319110553):

It's finally time to make some space to my old junk storage. I found two of my previus PCs there, which finally need to go to SER.

When reading this, please understand that I do own other PCs too. Also, I have owned more PCs than I currently run and these two heading out, its just that I don't have those ones anymore to write a blog post about.

New PC - 2018 - Reference

Just to get the difference, I'll post couple bad still images from my new PC:

Both pictures are actually crappy for a number of reasons. First, a Noctua NH-D15 CPU-cooler is huge. It weighs 1.3 kg and is 16 cm tall measured from top of CPU to top of Noctua's fans. It takes 16 x 15 x 15 cm space from the PC case, just to make sure there is proper air flow for the heatpipes coming from the CPU. For photography, it means everything else is hidden by it. Ufffff!

Then Asus motherboard has an aluminium shielding covering everything. That's most likely for EMC protection and heat management. In a properly cooled PC-case, black aluminium transfers heat out of the motherboard faster. Also, a Fractal Design PC-case is pitch black. What I have here is a black picture of blackess on black.

Finally, also the graphics card is quite a beast. It takes 30 cm in lenght and is over 5 cm thick. Yet again covering everything inside the PC, that would be worth looking at. So, this 2018 PC is visually quite boring. Maybe that's why people love having some sort of light show inside their transparent case. Ufffff! My case doesn't have any transparent parts in it, no PC-disco for me, thanks.

PC 1 - Pentium 4 from years 2002-2008

This one I used in couple of hardware configurations for many many years. Mostly with Windows XP.

After serving me well, I decommissioned this around end of year 2008. There are changed files in early 2009, but it looks like I haven't used this PC since. It has been mostly gathering dust in storage.

CPU

Spec:

• Intel® Pentium® 4 Processor 2.40 GHz, 512K Cache, 533 MHz FSB
• Launch: 2002/Q1
• Cores: 1
• 512KiB Cache
• 2.4 GHz
• 533 MHz FSB
• Socket 478

Details are at Ark Intel.com https://ark.intel.com/products/27438/Intel-Pentium-4-Processor-2-40-GHz-512K-Cache-533-MHz-FSB

It was my last PC not running 64-bit AMD64-instruction set at the end of 32-bit era Mighty powerful at the time, still less powerful than my iPhone 8 today. It looks like this with Intel's boxed CPU-cooler:

Board

The motherboard for this project is an Asus P4PE:

More details are at Asus' website: https://www.asus.com/supportonly/P4PE-XTE/HelpDesk_Manual/

Graphics card

Given the fact, that I never like putting enormous amounts of money into GPU's, so this one has a Club-3D Radeon 9800 PRO 128MB DDR GDDR:

The manufacturer's fan on top of the GPU was a piece of junk. I had to replace it in a year or so. Also, note the S-Video output between standard VGA D15 and DVI-D. Before HDMI, hooking up a PC to an everyday TV was tricky. S-Video was one of the supported connectors in TVs having multiple SCART-connectors.

Also looking at the pics of the CPU-cooler and the graphics card make me laugh. There is almost no cooling at all in neither. Also the simpliness of graphics card is something that really sticks out. Modern cooling, especially in graphics cards, looks HUGE!

One notable thing to mention, is that this graphics card connected to the MoBo via AGP or Accelerated Graphics Port and having DirectX 9.0 hardware acceleration making it pretty fast (at the time). If you've never heard of AGP, don't worry, it was just an attempt to make GPUs connect to CPU/RAM-package faster. It was little bit faster than PCI, but not much. Obviously, it wasn't such a great invention and was quickly surpassed by PCIe, which seems to stand the test of time. For curious ones, I googled up some specs for the card: https://icecat.biz/en-in/p/club3d/cga-p988tvd/graphics-cards-RADEON+9800+PRO+128MB+DDR-113233.html

PC 2 - Pentium from years 1994-1998

Ok, this baby is an old one! To make it really ancient, it even runs IBM OS/2 as operating system. Those padawans who don't know what an OS/2 is, just hit to Wikipedia for https://en.wikipedia.org/wiki/OS/2. During the active years in servie, I recall upgrading the CPU once, doubling the RAM once and swapping the GPU-card to a faster one. This is the oldest hardware I have, and didn't create any records at the time. For anything newer, I have proper records of the history in my wiki.

CPU

Spec:

• Intel® Pentium® Processor 150 MHz, 60 MHz FSB
• Launch: 1996/01
• Cores: 1
• 8 + 8 KiB Cache
• 150 MHz
• 60 MHz FSB
• Socket 7

Details are at Ark Intel.com https://ark.intel.com/products/49958/Intel-Pentium-Processor-150-MHz-60-MHz-FSB

This old puppy has one distinguishing feature: it is Spectre/Meltdown-proof! As it doesn't have any kind of prediction or jump analysis in it, it cannot be fooled. Actually, there is nothing in it to make it run faster. It is a product of an era when everything was made faster by adding megaherz (note: not gigaherz) to base frequency.

The installed CPU with Intel's boxed cooler looks like this:

Please, notice how a heatsink and a fan on top of it are almost 20 mm high!

Also, don't confuse this Pentium to any of the modern day CPUs Intel calls Pentium. This one was really one of the first ever manufactured Pentiums. Initially this had a slower Pentium in it (I think 90 MHz), just getting an upgrade was inexpensive at the time.

Board

Motherboard is a Asus P/I-P55T2P4. Some specs of it can be found from: https://www.asus.com/supportonly/PI-P55T2P4/HelpDesk_CPU

Layout of the board is:

Note how this is clearly a PCI-era board, but it still has three ISA-slots in it. And please, don't confuse PCI for a modern PCI-express. That's probably why, the word "conventional" is used in Wikipedia article of PCI: https://en.wikipedia.org/wiki/Conventional_PCI

GPU-card

Oh, this is a trip to memory lane. Card is a S3 Trio64V+. Some information about this relic is at https://en.wikipedia.org/wiki/S3_Trio and http://www.vgamuseum.info/index.php/cpu/item/359-s3-trio64v.

The card looks like this:

Note how I have Scotch Tape on two holes of the mounting bracket. Reason is simple, back in the 1990s we didn't know how air in a PC case should flow and we had it wrong. Sucking air trough graphics cards (or any other necessary components) is not smart and that's why its done differently today.

This particular card first appeared in 1995 and I've been using it for couple of years around that time. S3 was a really successful graphics company and their products were really good at the time. Also, at the time companies did design and manufacture their own cards. Today manufacturers just release specs and reference designs for the actual manufacturers to do the heavy lifting. Also, for old geezers like me, it is soooo weird to see a graphics card to be running just-the-regular PCI.

For brief history of GPUs in the 90s, read something like From Voodoo to GeForce: The Awesome History of 3D Graphics @ PC Gamer. Just like CPUs, also GPUs have gone giant leaps during past 20 years.

If you've never heard of company called S3 Graphics, go educate your self at https://en.wikipedia.org/wiki/S3_Graphics. Brief highlights are January 1989, founding of the company, November 2000 at point where S3 had officially been outran by competition, they changed their name to SONICBlue. And finally March 2003, when filed for Chapter 11 bankruptcy. They had their moment, but given the fierce competition by Nvidia and Ati and Matrox and many others, they just couldn't keep up with the technological advances fast enough to be able to offer interesting products to customers. Today, what remains of S3 is part of HTC, the Chinese cellphone manufacturer.

Ok, the junk's gone - What then?

Yes, it cleaing up doesn't end here. I still have ~1000 floppy discs in storage. Now that PCs having a 5.25" or 3.5" drive are gone, I'll keep cleaning some more. Maybe I should take a peek into some of those floppies to see if there is anything valuable in them.

Also, I have the images of the OS/2 hard drives. Next I need to figure how to read them.

This is a weird one. Beginning of this year Qnap made a choice to EOL my NAS-box. I posted about that at the time. Well, that happens. Its just that my current box works for me and all, I don't necessarily need a new one.

Without announcing anything, I got a firmware upgrade for it! Actually, I got two. Initially I assumed to be hallucinating or something. On a 2nd time I had to confirm my original blog post, that EOL really did happen.

Today on their EOL-page (or Product Support Status) https://www.qnap.com/en/product/eol.php they state:

Model: TS-419U II
Hardware Repair or Replacement: Full
OS and Firmware Updates and Maintenance: 2017-12 (QTS 4.3.3)
Technical Support and Security Updates: 2020-12

What the ... happened? Now they're announcing (or NOT announcing, informing) to provide full support until end of year 2020. Well, maybe I should go purchase a new one anyway.

Today, I'm combining to previous post into a new one. I've written earlier about going fully Let's Encrypt and the problems I have with them. Now that I'm using them, and those certs have ridiculously short life-span, I need to keep automating all possible updates. That would include the IPMI 2.0 interface on my Supermicro SuperServer.

Since Aten, the manufacturer of the IPMI-chip chose not to make the upload of a new certificate automateable (is that a word?), I had to improvise something. I chose to emulate web browser in a simple Python-script doing first the user login via HTTP-interface, and then upload the new X.509 certificate and the appropriate private key for it. Finally the IPMI BMC will be rebooted. Now its automated!

So, the resulting script is at https://gist.github.com/HQJaTu/963db9af49d789d074ab63f52061a951. Go get it!

Roughly 4 years ago, I blogged about a battery change to my UPS. The post is at APC Smart-UPS battery change. My unit eats APC Replacement Battery Cartridge #7 as replacment, and they are generally available in the net. The price point is there, such a replacement costs 250,- € easily. Much more, if you're not careful.

Couple years after publication Mr. Oliver commented my post (https://blog.hqcodeshop.fi/archives/195-APC-Smart-UPS-battery-change.html#c2000) about getting a pair of Yasa NP7-12 batteries. In his comment, he posted a PDF-spec http://www.yuasabatteries.com/pdfs/NP_7_12_DataSheet.pdf. Just by eyeballing the details, it became obvious, that there is no way in freezing hell, to be able to use that particular battery unit as replacment.

While I dismissed the suggestion quickly, Mr. Oliver succesfully incepted the idea (if somebody hasn't seen Inception, you missed my point there). In life, there are situations where the plan is a crappy one to begin with. On the other hand, sometimes the plan is rock solid, but implementation falls short. For best results, a good plan and good implementation is needed. So, I decided to investigate this battery replacement thingie and come up with a good plan. Initially it was more like a wish, I had no way of knowing how my chips would fall out.

The Investigation

So, when yanked out of the UPS, a APC Replacement Battery Cartridge #7 looks like this:

During the 4 years of running, it gathered some amount of dust. If I would care, I would have cleaned the worn out unit before taking the pics, but ... naah. And if you want to know how to actually yank it out, see my previous post.

In a glance, the #7 doesn't have any moving parts in it. There is nothing to remove, nothing to un-screw. But a closer inspection reveals some plastic covers just attached to the battery with a two-sided tape:

Yes! I'm, getting somewhere here. A close-up on the battery connectors:

The battery connectors have holes in them and there is a M6-screw running trough them. A 10mm wrench and a PH2 screwdriver will do magic there.

Finally I had all the parts separated:

There was some adhesive tape to make the two batteries stick together. As all the connector bits were removed, I just applied brute force to separate the lead acid batteries from each other.

The Plan

Battery used by APC:

A Hitachi Chemical Energy Technology Co. Ltd, GP12170. Spec is at: http://www.csb-battery.com/english/01_product/02_detail.php?fid=5&pid=13

My simple plan was to:

1. Find out if a suitable replacement battery was available. Mr. Oliver suggested that the price range would be £30,-
2. Get the replacement batteries
3. Apply some adhesive tape and screw the APC-connector bits and their plastic covers back
4. Plug the refurbished unit back to my UPS and admire the results (success of failure)

The Implementation

Finding and getting the replacment units

Nope. Just by googling, I didn't find that particular GP 12170 battery anywhere where the shipping costs wouldn't kill me. Lead acid batteries are heavy, as in expensive to ship, remember, the lead-part there.

• Since asking doesn't hurt, I just popped by my local battery-guy at Akku-Arkka Oy.
• His first question was: "Which lawnmower did your take that from?"
• I was in luck! He had suitable units in stock. For some reason, they are sold as a twin-box:
  
• Obviously, a twin-box is exactly what I needed for this purpose!

Assembly

At this point, my plan was coming together.

1. I just got some two-sided tape, stuck some of that on the side of the battery and stuck the other battery to the tape to form a single unit.
2. I screwed the APC-bits back to the connectors. Even the holes were precisely the same size.
3. More two-sided tape to the top and battery connectors were nicely covered.

I didn't bother taking any pics of this. My final result looked un-surprisingly like the original APC-unit.

Plugging it in & testing

Since these quality UPS-things have hot-swappable batteries, the UPS-unit was running my computers all the time since the batteries failed, I removed the old battery-pack and finally was about to test the new battery-pack. The obvious risk at this point was if I made a mistake and my UPS would completely fry because of that.

But no, it didn't happen. Everything worked perfectly! My APC utilities on Linux indicated following:

# apcaccess
APC      : 001,043,1009
DATE     : 2018-05-20 13:03:11 +0300
VERSION  : 3.14.14 (31 May 2016) redhat
CABLE    : USB Cable
DRIVER   : USB UPS Driver
UPSMODE  : Stand Alone
STARTTIME: 2018-05-20 13:03:07 +0300
MODEL    : Smart-UPS 1500
STATUS   : ONLINE
LINEV    : 234.7 Volts
LOADPCT  : 13.6 Percent
BCHARGE  : 100.0 Percent
TIMELEFT : 91.0 Minutes

Finally

Looks like all the lead-acid batteries in world come from Vietnam. See article Is Vietnam the new China for lead-acid battery manufacturers? about that.

I saved ~150,- € by doing this instead of going for the official unit. Nice!

Whenever there are new toys, I'm excited! Now that I have new iPhone 8, there is no other way to phrase it: it IS same as 7, which was same as 6. The observation Woz pointed out when X and 8 came out:

I’m happy with my iPhone 8 — which is the same as the iPhone 7, which is the same as the iPhone 6, to me.

Here are the comparison pics:

Last year around this time, I wasn't especially impressed when I got my 7. See and/or read it at /archives/345-New-toys-iPhone-7.html. This year, I'm kinda hoping to still have my 6. The good part is, that I won't have to pay for these toys myself. If I would, I would be really really disappointed.