Stuart

Just come across this, very much in the style of the 80s. http://www.telegraph.co.uk/technology/0/the-science-behind-how-the-internet-works/?playlist=series%3Ahow-it-works There's a series of videos all done in the same way covering batteries, GPS, mobile phones, ...

There was a project to download off the screen as well. The only reference I can find to it is from http://sleepingelephant.com/ipw-web/bulletin/bb/viewtopic.php?t=5875: "I remember on the BBC there was a computer programme where they experimented with transmitting code while the show was broadcast. There was a hardware project to build a light dependent resistor into a suction cup and some circuitry to decode this to the serial port. You placed this suction cup over the top right corner of the tv and black and white squares were flashed during the show. After that we also had magazines giving away Flexidisc 45rpm single records to save typing in."

Indeed, We even had program downloads over the TV during the programme at one time. Can't remember which micro it was for specifically, but you had a light-dependent resistor(?) in a little black suction cup that you stuck over the corner of the TV picture and plugged the cable into the micro tape(?) input and typed "load". They broadcast the programme with a flashing block in the corner of the picture that downloaded the program to the micro.

It's gone to Jim. The woman put it on Craigslist just after she contacted me. It was still on Craigslist for a while after the deal was done with Jim.

If you take a look at my port of "The Valley" (http://www.stuartconner.me.uk/ti/ti.htm#the_valley), that runs on standard TI XB and uses some assembly routines to implement the 40 column mode. Once switched to 40 col mode, all screen printing is done through CALL LINK() routines. There's also Cortex BASIC (http://www.stuartconner.me.uk/ti/ti.htm#programs), which runs in 40 col mode by default.

The LED only comes on when the 32K is accessed. So you won't know if it's working just by plugging power plug in. A few bent pins on the expansion connector won't matter as long as they as they're not touching each other (or anything else).

This looks like them: http://www.arieselec.com/products/data/12024-dip-collet-solder-tail-socket.htm. Datasheet suggests that the latches *should* lock the IC into place?

Are they not a normal turned-pin socket, but with an eject mechanism?

And just to clarify - this is TM990 stuff - the industrial modules - not TI990 - the office system. Understood?

Just sent the message.

I should have placed a bet that you would be first, Jim! I'll add you to the e-mail thread in a minute using your ksarul e-mail address in your message. Let me know on here if nothing arrives. (And I think it's a woman - Alexis.) EDIT - saw the e-mail address in the message!

To take your minds off the US elections, I've had an e-mail from someone in Oregon, US who has got some TM990 module/parts for sale. They appear to be brand new, unused, boxed, with manuals (boxes have just been opened to verify the contents). Together they'll form a nice little minimal system that will run the TI Power BASIC software, plus the TIBUG monitor software, plus other stuff. Components are: -- 4 slot card cage -- TM990/101 Processor module -- TM990/302 Software Development Module -- TM990/201 Memory Expansion Module (I need to double-check that there is one of these) -- Set of Power BASIC EPROMs The seller is after $150 for the lot, plus postage/packing costs (each component is around 2-4 lbs. with the manual). To get it working, you'll need to be able to hook up a power supply to the card cage power terminals. A small PC power supply will work fine but you'll need to be able to solder some solder tags onto the power supply cables and sort out a power switch. You'll need a serial cable to connect the Processor module to your PC/laptop/whatever running a serial terminal application. Although the components are brand new, it's all over 30 years old so no guarantees it will work (you might need to replace a RAM chip or two for example). Having said that, the manuals have full schematics and all the chips can still be obtained, and I'm happy to help out with any problems (and have duplicates of the EPROMs). I've got spares of the TM990 modules that I've designed that I'm happy to negotiate on separately - IDE interface module (connects a hard disk or compact flash, with a simple disk operating system), 64K battery-backed RAM module, VDP and sound module (designed to use a TMS9918 with composite video output, but also works with an F18A but you'll need to build a small adaptor to reposition the IC socket). Details of the TM990 modules can be found here: http://www.stuartconner.me.uk/tm990/tm990.htm Some pictures of the bits actually for sale: https://portland.craigslist.org/wsc/sop/5865835012.html (it's on Craigslist for $200, but I've arranged to buy a second processor module separately, so $150 for the rest). If you're interested, PM me an e-mail address that I can pass on to the seller. The two of you can then come to whatever arrangement you like (no comeback on me should anything go wrong!) Stuart.

I believe the TI transformer gives 16V and 8V - and the transformer on that link doesn't have 8V. Surely there are enough old power bricks kicking around that you could take the transformer out of one of those, if you're tinkering? Then you know you have exactly the right one.

I've updated the web browser so it now supports URLs specified by IP address rather than host name. Version 9.2 available on my web site: http://www.stuartconner.me.uk/ti/ti.htm#internet_web_browser. =============================== *Version 9.2 changes: *-- Added support for a URL in the form of a numberic IP address, in addition to a textual host name. So the * following example web page URL formats are now supported: * -- www.stuartconner.me.uk/tibrowser/z.htm * -- myti99.com/ * -- 198.136.54.70/~stuartc2/z.htm * The URL in a hyperlink on a page can specify either a host name or an IP address.

One reason - the further the speech synth is from the side of the console, the more interference you'll get on the audio. Just putting it on one of Ksarul's side port splitters and there is a noticeable buzz on the audio. (Do the side port devices like the TI 32K memory and disk controller pass through the audio input line?)

No, the page is stored in the download buffer exactly as received, so no sloppiness! At 38,400 Baud and with no handshaking, there's not really enough time to do any processing of the page data as it is downloaded.

Well done Erik! Quite amazing what you've done there in just a month!

The processor is not getting far enough through the 'boot code' to silence the sound chip and initialise the screen. As others have said above, there are any number of failure points that can cause this. One easy thing you might want to try is to remove and refit the set of three socketed GROM chips and the sound chip, just in case they're making bad connections in the sockets after all these years. Also try swapping the PSU board and power brick for known working ones, see if that makes a difference. If that doesn't help, then its out with the schematics and test equipment ...

So where's the PIO port on a NanoPEB???

So it sounds as if the best approach is to determine *in the browser code* whether an IP address has been entered, and if so there is no need to call the DNS PHP script? Then you could use it on an internal network that isn't connected to the Internet. Corey, any idea how big the largest browser page you have is? Interested to see how many bytes I can grab from the page buffer to use for code, if needed.

Thanks Corey. I actually need to park this for a couple of days until I can look at the browser source code (I'm away from home at the moment). I've a vague memory that the "www." prefix is required/expected in some areas of the program, which would give a problem if using an IP address. I need to check that before playing with the PHP script.

Any disadvantage to using ... $valid = filter_var($string, FILTER_VALIDATE_IP); ... instead of a regex?

Could you conjure the PHP line needed to determine if its an IP address rather than a host name?

It will treat the IP address as a host name and try to convert it to an IP address (which with the current PHP script will fail), as ElectricLab says. If the have the means to hex-edit the browser disk or EPROM image you're using, it should be possible to modify the program to point to your own server to do the DNS conversion - but you'll need details of the syntax the browser is using/expecting. I'll try to sort out some details over the weekend.

So from that data it looks like a short between the two bits marked with an X: – – – X – – X –? Trying checking at the EPROM sockets for a dead short between each data pin and each of the other data pins. Possibly a short from a pin to a trace running next to it under the socket. (If you end up removing then refitting the sockets to cure the problem, see if you can raise them up a millimetre or two so you can see below the socket body where the traces go between the pins.)