Picked up a Cortex yesterday!

[Ash Ward]

Hi all,

 

I managed to get hold of this beauty yesterday and was directed to this forum.

 

I know very little about TI based computers in general, but this thing looks awesome. Apparently it's working, but I want to give it a once-over first (check caps, voltages, etc.) before I fire it up and haven't had a chance yet.

 

I haven't opened it yet, but it hasn't got any drives in it so I guess there's no disk controller either. It's also missing the cassette cable so I'll need to get hold of the schematics and work out the pinout to make one up.

 

It seems to be in excellent condition. The guy I bought it from said he built it himself from new, and it looks like it's been really well looked after.

 

It also came with a handful of tapes. A few of them are homemade, but there are four interesting commercial ones (image attached) - I'm loving the hand drawn inlays!

 

It also came with the TV (also in great working condition and with the very chunky remote!) - I imagine it was the same TV he originally used for it.

 

Anyway, I'll update this thread as I delve into it. Also any extra information would be greatly appreciated because other than the excellent http://www.powertrancortex.com/ site, there's not much other info available online!

[+dhe]

Congrats - awesome shape!

[+Ksarul]

There are three excellent resources online for the Cortex. You already found the most comprehensive one from a documentation/software standpoint. In addition, there is information on a mini-Cortex out there by @Stuart. Lastly, there is a huge Cortex thread on the VCF forum that I started back in 2009 when there was really nothing online about the Cortex. It even includes some interesting background information from the guys who developed the Cortex at TI. Several of the folks here have Cortices, so you are not alone. The tapes are interesting. They look like they are all from Andy Paddon software. There are images of the tapes online (on the WHT site, IIRC), but yours are the first time I've seen the inserts. It would be good to make images of those to put onto the Cortex website.

 

On disk drives, the original controller chip is mostly unobtainium (TMS9909). Original chips show up once in a great while, but most of the ones seen for sale are inoperative fakes. There is documentation on the Cortex website to show how to use a different chip (2797) instead. Most folks with disks went that route.

[Ash Ward]

Thanks @Ksarul - that's really useful. I'll have a read through your thread - some nice light reading for this evening! ?

 

I spotted the 2797 based disk controller schematics. I'll definitely have a go at making a couple as I do fancy adding the drives to it at some point - plus I'd like to learn more about floppy disk interfaces too. It's probably easy enough to make them on prototype board, but is there a laid out PCB design available?

 

Is there anything to watch out for with these machines? I've had a quick look through the schematics and the power supply looks nice and simple so probably not much to go too wrong, but is it worth putting in a modern switching supply? What are the chips like for heat management - is it worth adding some heat sinks? I notice that mine doesn't have a fan in but there's space for one on the back of the case - is it worth adding one? Are there any chips in there that are particularly difficult to replace that I should take extra precautions with?

 

I noticed the RGB board schematics too - might have a go at that at some point too. In the mean time I assume a composite mod would be relatively simple?

 

Apologies for all the questions (and it may be that many of them are already answered in your VCF thread), but I'm excited to have something rather interesting to learn about and work on!

[+Ksarul]

No one has done a 2797 daughter-board yet, although it would be a very simple thing to do (and it is on my to do list). Most of the other chips on the board are relatively easy to source if you need replacements. The 74LS612 is probably the most difficult, and even that one is somewhat easy to source (beware of some of the Chinese counterfeits though). One other chip in the mix is unobtainium: the 74LS2001. The chips were never readily available to anyone outside of TI. It would be possible to use the diagrams we have to make an FPGA replacement for it--or for simpler uses, there is a viable work around for it included in the ETI articles.

 

 

[Stuart]

Welcome, Ash. Presumably you're in the UK? I'm down in Devon, if that is anywhere close to you.

 

Was the Cortex advertised somewhere, or did you do a bit of private negotiation?    I've got a Cortex-in-a-case that I added floppies to, and a bare PCB that I built up.

 

The Cortex-in-a-case had a power supply problem when I got it - one of the big caps kept blowing one of the diodes. Replaced cap and diode and all good.

 

Where have you seen the 2797 schematic? I've seen a parts list for the PCB, but don't have a copy of the schematic in my notes, and have never seen a PCB layout. You may see the original TMX9909 FDC on eBay from China - don't be tempted - a few of us have tried and you either get a 40-pin chip that is not a TMX9909, or you receive nothing and the order is eventually cancelled.

 

You've probably found my website [http://www.stuartconner.me.uk/powertran_cortex/powertran_cortex.htm] - a few useful bits on there.

 

 

[Ash Ward]

Thanks @Stuart

 

I actually spotted it on Facebook Marketplace. Thought it looked interesting so went and bought it!

 

I'm in East Sussex, so not particularly close to Devon unfortunately.

 

I was wondering about the power supply and whether it would be safer to use a modern switching one (although the old linear supplies should be pretty reliable as long as they don't get too hot!)

 

I was talking about this schematic on the powertrancortex.com website for the 2797 based controller card http://www.powertrancortex.com/hardware/TMS2797_Replacement_Disk_Controller.pdf which I assumed was what I might need. I haven't looked in much detail though so may be mistaken!

 

@Ksarul mentioned about the game tapes, so I've scanned in the inserts and tape stickers in case anyone is interested. The stickers fell off of two of the tapes so I've got a good scan of those, but one was still well attached so I've just scanned the tape itself for the moment. I'll see if I can get a better scan at some point.

 

There were also a few home tapes that came with it. I'm not sure if they're known or unknown games, or just basic programs the guy had typed in. There's "Vaders" (which I assume is some Space Invaders clone), "Nibblers" (which is probably a snake clone), and "Haunted" (which I have no idea, unless it's the haunted house game from that Usborne book which I remember typing into my Commodore 16 back in the day!)

 

 

[Ash Ward]

I didn't have much time, but did manage to have a have a quick look inside it last night. It looks nice and clean and well put together. I noticed that mine doesn't have the DMA controller or MMU populated. Should I be looking to add these (and associated components) in, or is there nothing really to gain? It seems the chips are readily available and relatively inexpensive.

 

Date codes all look like late '82 - early '83, so I assume the kit was produced in early '83.

 

I also spotted that the 5-pin DIN Cassette port on the back isn't actually wired to anything! I assume the unmarked PCB mounted 5-pin DIN connector on the motherboard does the same thing? Does anyone have the pinout for it? (The ETI articles seem to be lacking rather a lot of info - I assume there must have been more comprehensive instructions with the kit?)

[Stuart]

Ash, looks like your PCB is fitted further to the left than usual, so the cassette and RS-232 ports are accessed directly through the side of the case rather than being wired to the connectors at the rear. Other Cortex PCBs are usually fitted further to the right so the expansion connector is up against the side of the case. Wonder if that was a late factory option?

 

I think you need the DMA controller for floppies. Only need the MMU for external expansion, or just the joy of fitting it. Note that there are a couple of links (tracks) on the PCB you need to cut if you fit the MMU.

[Stuart]

Normally have the speaker fitted over the fan outlet by the way. No fan needed as all the hot PSU regulator are fitted directly to the case. Can use a PC PSU but then you lose the smell of hot vintage electronics. ?

[+FarmerPotato]

@Ash Ward That is an amazing looking machine! I will be following along, I hope you get it running!
 

[Ash Ward]

Work and family are keeping me pretty busy, but hopefully I'll get some time to test this thing out soon and let you know what kind of state it's in functionality-wise!

 

I want to try out some software on it too but I have no cassette cable. Until I can get the stuff to build one (and figure out the pinout - is it the same as the Amstrad CPC, Acorn, or MSX pinouts I wonder?) I was wondering if firing stuff across the RS232 port might work (I have a USB <-> RS232 adapter somewhere). I assume it's proper RS232 level (+12V/-12V) rather than TTL (0V/5V) or something? What commands might I need to load something in, and are there any binary images anywhere I can try out?

 

I'm also planning in my head some of the extra things I want to do. Implementing the missing floppy controller is probably the first big thing. Rather than actual floppy drives though, I was considering using something like the GoTek with flashfloppy. I can't see why it wouldn't work, but has anyone tried that on this machine yet?

 

Regarding the FDC, is there a list anywhere of the components I need to populate to add the DMA controller?

 

Also I noticed that there are both TMS2797 and WD2797 chips available. Are the TI or Western Digital ones better or about the same? If there's no real difference then I'll get a TI one for the authenticity, but I just wondered if the WD might be more reliable?

[Ash Ward]

I've actually just spotted that @Stuart has instructions (and images) on his site for loading stuff in via the RS232 port which is great! I shall give that a go!

[+Ksarul]

The ETI articles have the chips listed to populate the DMA controller (I think it was in the third article). On the manuals from Powertran, they were literally just a reformatted version of the build articles in ETI, so you won't find much new information there. One of the three original ETI articles also has the assembly instructions for the cassette cable (it was either the second or third one, IIRC).

[Ash Ward]

So I had a bit of time yesterday to test it out. Disconnected the power connector from the main board and tested the voltages, which were all spot on, so I connected it all up and fired it up.

 

Unfortunately my LCD TV didn't seem to be able to sync properly, but the picture was good enough to see that it was working and it booted into BASIC. At first I thought it was somewhat broken though as every command I typed resulted in an error, but I then realised that the commands need to be in upper case (not surprising), but rather infuriatingly whoever designed the character set decided to have small upper case characters instead of lower case ones so I didn't realise I was typing stuff in lower case! Is that character set the same across the whole TI range (as it's a bit weird!). Anyway, once I'd realised that, I managed to get some simple programs working.

 

However, the poor sync on the LCD was annoying (screen rolling every second or so), so I tried it on the CRT TV that I got with it. That worked fine, but I'd honestly forgotten how flickery CRT screens are (were they always that bad, or has the old phosphor lost some persistence or something?)! Anyway, I couldn't look at it for more than a minute or so and had to give up!

 

So the good news is that the Cortex looks like it works beautifully. The bad news is that I don't currently have a suitable screen for it that I can stand to use for more than a few minutes!

 

That means that I think the first hardware project for it now going to be to sort out a video solution - either building the RGB interface (which will hopefully work better with the LCD TV than the UHF does), or maybe just hook an off-the-shelf YPbPr to HDMI converter off of the output (with a level shift) from the TMS9929 to go straight to HDMI?

[Stuart]

Keep an eye open for an F18A if you want to connect a VGA monitor. They work in the Cortex. See the F18A threads ... sounded like someone was going to make another batch.

[JB]

1 hour ago, Ash Ward said:

Is that character set the same across the whole TI range (as it's a bit weird!).

If I recall, it is the example character set in the 9918 et al documentation. So it could be considered the official character set of TI graphics chips.

 

1 hour ago, Ash Ward said:

I'd honestly forgotten how flickery CRT screens are (were they always that bad, or has the old phosphor lost some persistence or something?)!

I don't believe phosphors lose persistence with age, just brightness.

My understanding is european sets WERE always that bad. 50 Hz TV is just gross.

[Stuart]

15 minutes ago, JB said:

50 Hz TV is just gross.

Ouch! That hurt! Didn't you NTSC guys needs a tint control on your sets to correct the colour?    

[+mizapf]

It's been much too long since we had the last NTSC vs. PAL discussion.

 

(I mean, we know the better system, don't we?)

[+pixelpedant]

Especially in a 9918 context like this one, where PAL users got YPbPr coming off of the chip, while NTSC users just got NTSC Composite.  Advantage: PAL, for sure.   

[JB]

9 hours ago, Stuart said:

Ouch! That hurt! Didn't you NTSC guys needs a tint control on your sets to correct the colour?    

I mean, in the 60s, yes.

But the bizarro world I wandered into when I had to get an australian to explain scan-doubled TVs, then explain to him why nothing of the sort existed in my part of the world...

 

I'll take slight color impurities in a broadcast environment over headache-inducing flicker.

 

Also, our game consoles connected to channel 2, 3, or 4, not ... 37 or whatever. In VHF where they belonged!