27513 and Master ROM cartridge

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
Boydie
Posts: 117
Joined: Sat Oct 24, 2015 8:25 am
Location: Sunny Wigan

27513 and Master ROM cartridge

Postby Boydie » Thu Sep 28, 2017 1:46 pm

The Master ROM cartridge states that it can accept 2 x 27513 eproms (128k).
From what I've been able to find out, the 27513 is a 64k eprom which is divided into 4x16k pages in order to be more useful in an 8-bit system.

Has anybody actually tried using one (or more) of these eproms in a cartridge? How did the Master cope with having 4 rom images in a single slot - how did it select between them?

Or should the label have read "Accepts 2x27128 (32k) or 2x27513 (really expensive 32k)"?

crj
Posts: 145
Joined: Thu May 02, 2013 4:58 pm

Re: 27513 and Master ROM cartridge

Postby crj » Thu Sep 28, 2017 7:48 pm

I've just tracked down the datasheet.

It does indeed look a lot like four paged 27128s. The way you select one of the four banks is simplicity itself: you just write to any address in the EPROM. Bottom two bits of the value you write are the page number.

The ~WE line is on pin 27, which other devices might use either for programming or as A14. BBC Master cartridges have a write enable signal (because 1MHz bus is amongst the many other things they support) and looking at the schematic I see it's plumbed through.

The bad news is that, so far as I'm aware, the OS offers no native support for such paged EPROMs. In any case, the overall limit of 16 paged ROMs is unequivocally baked into the API. You could use it to software-switch between different ROMs, I guess. Or people could have used it to do a cheaper and tidier version of the hack Computer Concepts used for InterWord and SpellCheck.

So not completely useless.

Boydie
Posts: 117
Joined: Sat Oct 24, 2015 8:25 am
Location: Sunny Wigan

Re: 27513 and Master ROM cartridge

Postby Boydie » Fri Sep 29, 2017 7:18 am

That was what I assumed, too. What made me challenge the 16-rom limit was an old Morley ad for the Master AA board, in which they claim on-board support for an extra 8x16k roms, plus 4x32k, without interfering with the cartridge roms or sideways ram capabilities. The ad specifies "a total of 32 rom images simultaneously present in the machine, all available at the same time".

Hence my wondering, was this a bodge uniquely designed by Morley, or were they the only people to bother to utilise a little-known feature already present in the OS? If the latter, was this feature how Acorn intended to get 4 images available per cartridge rom?

Also, when the M128 came out ( a few months before cartridges did), a lot of reviewers panicked and claimed the cartridges would *only* take 64k roms. Presumably they misunderstood what Acorn were saying about the slots, but it does suggest Acorn were claiming the ability to use 27513s in a useful way in their press releases rather than it just being a label misprint on the cartridge.

And some manufacturers suddenly started marketing eprom blowers for the BBC, specifically capable of programming 27513s, which would suggest a perceived utility of these devices. Or cynical marketing.

Hence my speculation - has anyone ever tried them to see what actually happens?

And how did Morley get 32 simultaneous rom images?
Last edited by Boydie on Fri Sep 29, 2017 7:29 am, edited 1 time in total.

User avatar
hoglet
Posts: 6372
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol

Re: 27513 and Master ROM cartridge

Postby hoglet » Fri Sep 29, 2017 7:23 am

Was this the advert?
Morley_Master_AA_Board.jpg

RobC has one of these:
viewtopic.php?p=78609#p78609

Sounds like a second. manual level of paging is added.

User avatar
daveejhitchins
Posts: 3629
Joined: Wed Jun 13, 2012 5:23 pm
Location: Newton Aycliffe, County Durham

Re: 27513 and Master ROM cartridge

Postby daveejhitchins » Fri Sep 29, 2017 7:31 am

crj wrote:It does indeed look a lot like four paged 27128s. The way you select one of the four banks is simplicity itself: you just write to any address in the EPROM. Bottom two bits of the value you write are the page number.
Beware! Some ROMs and maybe the OS write to each available sideways bank checking for RAM . . . This could, potentially, change the bank! You'd need a write protection mechanism if one isn't fitted!

Dave H :D
Parts: UM6502CE, GAL22V10D, GAL16V8D, AS6C62256A, TC514400AZ, WD1772, R6522, TMS27C512, AT28C256
Products: ARA II, ABR, ATI, AP6, MGC, AP5 . . .
For a price list, contact me at: Retro Hardware AT dave ej hitchins DOT plus DOT com

Boydie
Posts: 117
Joined: Sat Oct 24, 2015 8:25 am
Location: Sunny Wigan

Re: 27513 and Master ROM cartridge

Postby Boydie » Fri Sep 29, 2017 6:53 pm

That wasn't the ad I'd seen. It's actually a lot more informative! The one I'd seen was from the Morley ads in The Micro User, which gave an awful lot more hype, including such lines as "32 rom images available at once". Quite clearly that wasn't the case after all!

Interestingly, I've just found a flyer on 4corn for the ACP advanced ROM Adaptor II, which also cited "support for 27513 and 27011 devices (when used with the correct software) for which we are currently developing new products". Wonder if they ever came up with "the correct software"?

User avatar
1024MAK
Posts: 6679
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: 27513 and Master ROM cartridge

Postby 1024MAK » Fri Sep 29, 2017 10:05 pm

If a 27513 was used, whatever software was used would have had to run from RAM.

And 27513 EPROMs would not have been that easy for most users to get hold of I would have thought.

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

crj
Posts: 145
Joined: Thu May 02, 2013 4:58 pm

Re: 27513 and Master ROM cartridge

Postby crj » Sat Sep 30, 2017 12:10 am

Just for reference, here's the show-stopping kicker with the idea of ever having more than 16 simultaneous sideways ROMs.

Each ROM has a number, 0-15. Each ROM also has a single private byte of workspace at &DF0+ROM number. The intent is that ROMS which claim private workspace use that byte to store the base of their allocation, so they can find it again later.

If there were ever a ROM 16, its private byte would be at &E00, which... no.

The alternative would be to have multiple ROMs with the same number. But then your problem is what to do when one of them claimed an extended vector. And when one of them launches as the current language but you want the others to continue operating.

I just spent longer than I'd care to admit thinking really hacky thoughts about workarounds. But at the bare minimum they'd make each service call take about 0.2ms longer per doubly-sideways ROM, which is tantalisingly close to being out of the question. It was when I contemplated meta-paging a language ROM, but having a hardware intervention substitute RTS for its service call entry point when address &8003 was read that I begain to question my sanity and stopped.


Why do you want so many ROMs at once, anyway? (-8

crj
Posts: 145
Joined: Thu May 02, 2013 4:58 pm

Re: 27513 and Master ROM cartridge

Postby crj » Sat Sep 30, 2017 12:28 am

Ooh! Ooh!

If you can sit between the CPU and the memory subsystem, then you can overlay a series of pages over the 97 bytes of RAM between &D9F-&E00. And you can make the overlays write-sensitive so that a management system can cheaply know when that memory was written to.

If you're prepared to go to that length, maybe it's possible after all. Though, it must be emphasised, only by applying a third-millennium sledgehammer to a nut that couldn't realistically have been cracked in the eighties.

(If you also made &D00-&D60 independently pageable it might not even be impossible to cope with doubly-sideways ROMs that claimed NMIs.)

User avatar
daveejhitchins
Posts: 3629
Joined: Wed Jun 13, 2012 5:23 pm
Location: Newton Aycliffe, County Durham

Re: 27513 and Master ROM cartridge

Postby daveejhitchins » Sat Sep 30, 2017 6:00 am

Boydie wrote:Interestingly, I've just found a flyer on 4corn for the ACP advanced ROM Adaptor II, which also cited "support for 27513 and 27011 devices (when used with the correct software) for which we are currently developing new products". Wonder if they ever came up with "the correct software"?
Unfortunatly not, however :D we've more chance today to fix this . . .

Dave H :D
Parts: UM6502CE, GAL22V10D, GAL16V8D, AS6C62256A, TC514400AZ, WD1772, R6522, TMS27C512, AT28C256
Products: ARA II, ABR, ATI, AP6, MGC, AP5 . . .
For a price list, contact me at: Retro Hardware AT dave ej hitchins DOT plus DOT com

Boydie
Posts: 117
Joined: Sat Oct 24, 2015 8:25 am
Location: Sunny Wigan

Re: 27513 and Master ROM cartridge

Postby Boydie » Sat Sep 30, 2017 8:02 am

crj wrote:Why do you want so many ROMs at once, anyway? (-8


1) I've spent the last 20-something years faced with a label that said it could be done, whenever I've used my Master.

2) I've finally acquired a usb eprom programmer that can do in 15 seconds what my WE Adder wasn't keen on doing in 15 minutes (plus stuff if couldn't do at all).

3) Affordable 27513s can be tracked down these days

So I thought the time had come to find out if it was possible, or if I should just tipp-ex out that bit on the cartridge.

crj
Posts: 145
Joined: Thu May 02, 2013 4:58 pm

Re: 27513 and Master ROM cartridge

Postby crj » Sat Sep 30, 2017 10:47 am

What EPROM programmer do you have?

I probably still have my manky old Solidisk ones somewhere, but if there are cheap good alternatives on the market now...

Boydie
Posts: 117
Joined: Sat Oct 24, 2015 8:25 am
Location: Sunny Wigan

Re: 27513 and Master ROM cartridge

Postby Boydie » Sat Sep 30, 2017 6:55 pm

I got one of these:

https://www.amazon.co.uk/gp/aw/d/B071H5XGR7

It may be cheap Chinese tat, but at 35 quid it was worth taking a risk. There's also a version with adapters for different chip formats for about 20 quid more.
So far, it's worked brilliantly with a variety of 27 series, from 16 to 64k, from a wide range of manufacturers. A (very) few of my eproms weren't in its database, so it couldn't auto-set its programming parameters, but picking an alternative chip if the same vintage/specs did the trick.

User avatar
danielj
Posts: 5136
Joined: Thu Oct 02, 2008 4:51 pm
Location: Manchester

Re: 27513 and Master ROM cartridge

Postby danielj » Sat Sep 30, 2017 7:00 pm

Does it do the 21V Vpp ones?

d.

Boydie
Posts: 117
Joined: Sat Oct 24, 2015 8:25 am
Location: Sunny Wigan

Re: 27513 and Master ROM cartridge

Postby Boydie » Sat Sep 30, 2017 7:13 pm

Yup. Most of mine are *really* old (35+ years) and it works fine with them. Even the ones which consistently failed with the Adder. Its database, by manufacturer, then their part code, is pretty extensive even for that vintage - I've only found two that aren't in it (one's an Intel chip that google doesn't know much about either).
Also lets you load four separate 16k images into consecutive blocks in its buffer for programming 27512s (or you can program 16k blocks separately on different occasions).

crj
Posts: 145
Joined: Thu May 02, 2013 4:58 pm

Re: 27513 and Master ROM cartridge

Postby crj » Sat Sep 30, 2017 7:42 pm

I wouldn't say that was cheap Chinese tat. That's expensive Chinese tat. (-8

I was expecting you to say it was a fiver, maybe a tenner. So far as I'm aware, the only difficult bit of what that's doing is generating the programming voltage.

I wonder where I put my Solidisk one...

Meanwhile, I'm staggered to discover they still make EPROMs.


Return to “hardware”

Who is online

Users browsing this forum: No registered users and 8 guests