Stuart

Console power consumption: http://www.avjd51.dsl.pipex.com/ti/ti.htm#power_supply_consumption

Nope, two different systems! The poster has a TI990 (with an "I"), I have TM990 (with an "M"). TI990 was aimed at office-type systems, and the TM990 for industrial control sort of applications. Stuart.

Thanks Lee. Without a Minimem cartridge, that rather puts a spanner in the works trying to confirm how the DSR is reading. Although with the 74LS245 now in a socket, you could I suppose bend out the /EN pin and connect it to it's driving signal via a switch, and a pull-up resistor to +5V. Could then disable it until after power up.

Have you got an E/A cartridge? Try switching the console with E/A cartridge on first so it boots, then switch on the PEB. Then try my BASIC program above. Type in and try running just line 10 first - it *may* not work (and would save you some typing) - not sure if the E/A cartridge has to be able to detect the 32K memory at power up in order for it to run CALL INIT.

Also see what happens if you fit the card with the new 74LS245 removed. Does the console then boot? That might point to a problem with the EPROMs, and the console is reading code that is sending it to never-never land.

Try also replacing the 74LS125 next to the '245. This controls the READY output (amongst others) and on my FDC this was shorting READY to ground, hence hanging the console. [http://www.avjd51.dsl.pipex.com/ti/ti.htm#resurrecting_disk_system]

Just adding how this program works. (Really annoying that you can't edit your own posts!) Line 20 pokes into memory the code for the following, starting at address >3000: LI R12,>1100 'CRU base address for FDC card. SBO 0 'Switch on card. MOV @>4000,@>3020 'Copy word from DSR ROM at >4000 and store at >3020. MOV @>4002,@>3022 'Copy word from DSR ROM at >4002 and store at >3022. B *R11 'Return. Line 30 changes the pointer at >202A to >3F30 to add an extra entry to the REF/DEF table. See the bottom half of page 276 of the E/A manual for further details. Line 40 adds the program name "FDC<space><space><space>" and start address >3000 in the new entry in the REF/DEF table. See page 276 of the E/A manual again. Line 50 calls the code just poked into memory. Line 60 gets the values copied from the DSR ROM to memory at >3020 and >3021. (I've just realised that the code above is copying 4 bytes, whereas only meant to copy 2!) Found (on Classic99 at least) that CALL PEEK can't read from the DSR ROM directly as for some reason the routine is first switching the card off, hence the need for the code above to copy the bytes we need to another location, then the CALL PEEK can read them from there.

Assuming you have an E/A cartridge, the following will let you read the first couple of bytes of the FDC DSR, to test that that section of the circuit is working. Short enough to type in. Make sure you've got a GOOD 32K RAM fitted when doing this! 10CALL INIT 20CALL LOAD(12288,2,12,17,0,29,0,200,32,64,0,48,32,200,32,64,2,48,34,4,91) 30CALL LOAD(8234,63,48) 40CALL LOAD(16176,70,68,67,32,32,32,48,0) 50CALL LINK(“FDC”) 60CALL PEEK(12320,A,B) 70PRINT A 80PRINT B If the DSR can be read, it should print the values 170 and 2 (>AA, >02 - the first two bytes of the TI FDC DSR).

On the FDC, I'll put my money on the 74LS245 data buffer being bad. Have you got a Minimem cartridge? You can use that to switch on the card by writing a 1 to CRU address >1100, which should light the LED on the front of the card. Next look at memory locations >4000 onwards which should be the DSR on the card (starts >AA, >55 ISTR - someone with a card handy would need to check to be sure). That will prove that a lot of the card is working.

Personally I've always had more luck with smaller cards. Been using a 32*MB* card for years with no problems. Have multi-*GB* cards that work fine on the PC but not the nanopeb. Small cards still available cheap on eBay.

As the museum is Government-funded, they couldn't afford one at current eBay prices ...

TI-99 on the telly tonight on the 2015 Richard Dimbleby Lecture, which was filmed in the computer section of the Science Museum. A couple of vintage computers displayed on the wall.

Use the % operator. But you can't use it directly in a PRINT statement. So in your code above you'd need something like: 30 $B=%A 35 PRINT $B

You can save and load memory images to cassette. As well as BASIC programs.

(1) If you mean "does it use Direct Memory Access?" then yes. In as much as the floppy controller chip uses DMA to read/write direct to memory, without the involvement of the processor. (2) Bitmap mode - yes. It uses the same family of VDP as the TI-99. Plus bitmap mode support direct in BASIC. (3) Sprite coincidence and automotion. Functionality not built into BASIC, from what I remember. But from BASIC you can access the VDP directly and check for sprite coincidence and do sprite movement in a loop. 10 COLOUR 1,0 20 GRAPH 30 COLOUR 40 A=01038H 50 B=0547CH 60 C=09244H 70 D=04482H 80 NME=42 90 SHAPE NME,A,B,C,D 100 MAG 1,0 110 C=90 120 D=80 130 FOR X=0 TO 8*3.1 STEP 0.1 140 SPRITE 0,X*10,C+D*SIN[X],NME,6 150 SPRITE 1,X*10+16,C+D*COS[X],NME,4 160 SPRITE 2,X*10+32,C+D*SIN[X],NME,10 170 SPRITE 3,X*10+48,C+D*COS[X],NME,12 180 NEXT X Your later post - string arrays. Completely different to TI BASIC. More like how arrays are handled in C, I believe. All explained in the Cortex User Guide.

When you say "The idea was brought to us by ..." does that imply you worked for Powertran at the time?

Take a look at post #163.

Hi, sounds like you need to familiarise yourself a little with the machine and processor architecture. You might want to take a look at [ftp://ftp.whtech.com/datasheets%20and%20manuals/Hardware/99-4A%20Console%20and%20peripheral%20Expansion%20System%20Technical%20Data.pdf] and [ftp://ftp.whtech.com/datasheets%20and%20manuals/Datasheets%20-%20TI/TMS9900_DataManual.pdf]. (In short, GROM is a type of memory accessed through a memory-mapped port, and CRU is a I/O interface unique to the TMS9900 family processors.)

The one with the DIP switch is a real-time clock plus 2K battery-backed RAM. It all looks rather nicely done.

Looking at the colour bars, it's a perfect match for the first of those screenshots, Michael. There - I told you they would be useful. The screenshot name is "vram 7f" - so which bit is 'stuck'?

If the connector is the IDC type, you should be able to carefully pull the individual wires off the connector. Then get a suitable length of normal ribbon cable. It will probably have two wires in the space for each connector position. So lay the end of the cable on flat surface, then using a knife cut 1/2 inch up between the wires. Then snip off every other wire and you've got a bit of ribbon cable with the same cable spacing as the connector. Fit to the connector - you'll probably need an IDC insertion tool - only a couple of $. Repeat at the other end and solder to the board. I used to have a cable that was a couple of feet long and it worked fine.

Converted the first (un-optimised) listing to Cortex BASIC and ran it on a couple of systems. Results below. -- TI-99/4A with a Cortex BASIC cartridge (running under Classic99, normal speed): 1 hour 10 minutes -- TM990 system (TMS9900, 3 MHz clock): 48 minutes -- Powertran Cortex (TMS9995, 3 MHz clock): 35 minutes -- TMS99110 breadboard system (4 MHz clock): 12 minutes Listing: 140 COLOUR 1,7: GRAPH 150 XP=144: XR=4.71238905: XF=XR/XP 160 FOR ZI=-64 TO 64 170 ZT=ZI*2.25: ZS=ZT*ZT 180 XL=INT[sQR[20736-ZS]+0.5] 190 FOR XI=0-XL TO XL 200 XT=SQR[XI*XI+ZS]*XF 210 YY=(SIN[XT]+SIN[XT*3]*0.4)*56 220 X1=XI+ZI+160: Y1=90-YY+ZI 235 PLOT X1*0.75,Y1*0.75 240 UNPLOT X1*0.75,Y1*0.75+1 TO X1*0.75,144 250 NEXT XI: NEXT ZI Stuart.

Already is. Take a read of [http://www.avjd51.dsl.pipex.com/ti/ti.htm#programs] for the details and EPROM data download. Have repaired the link to the Cortex user guide. Thanks for pointing it out. Stuart.

3 hours! Was it actually plugged in? I actually used this 'graph' as a benchmark for various configurations of TMS99xx processors running a port of the Powertran Cortex BASIC. Quickest was 1 minute 17 seconds. [OK, OK, I suspect yours is a little more detailed. And it's not symmetrical so you have to calculate both the left and right sides. ] Details at [http://www.avjd51.dsl.pipex.com/tms99110_breadboard/tms99110_breadboard.htm] and scroll right to the bottom of the page. A copy of the program for the BBC Micro here: [http://41j.com/blog/wp-content/uploads/2012/03/beebug_3dobj.jpg]. Slightly different parameters for the graph again.

Right, I've posted a new version of the browser - now up to version 6. http://www.avjd51.dsl.pipex.com/ti/ti.htm#internet_web_browser I've removed the 'M' key to display the previously entered URL which, on reflection, was perhaps not a good idea, and implemented a 5-entry URL history list. Any URL you type or select from a link is added to the top of the history list, with the oldest entry dropping off the bottom. If you select the same link twice in succession then the same URL is added twice to the history list - there is no filtering if a URL is already in the list. If an entered URL is invalid, it is also added to the list. To access the URL history list, press <FCTN-E> or <FCTN-X> (the up and down arrows) to cycle to the previous or next entry in the list. These keys are active both when viewing a web page and when entering a URL. The displayed entry can either be selected again immediately (press <Enter>) or edited. So if you type a URL or select it from the favourites list, after the page loads, pressing <FCTN-E> enables the page URL to be edited. Tested on my TM990, but have done only minimal testing on the TI-99 version using Win994a Stuart.