BBC B Elite versions

[rmbrowngr]

I came across a disk for the BBC with Elite and was surprised to find it had three versions of Elite e.g. Standard, Sideways RAM and second processor. I’ve heard of both standard and second processor versions, but not the version supporting sideways RAM. A quick search with Google didn’t bring any hits for this. What features does this version give?

[jms2]

No idea - never heard of it! You might have found an interesting lost relic there. I assume its not an official release disc?

[stephen_usher]

I think it's actually the "BBC Master" version that someone has renamed.

[rmbrowngr]

I think it's actually the "BBC Master" version that someone has renamed.

Not sure that's correct. I just opened the menu (BASIC) on the disk:

Code: Select all

 160 IF O%=0 PROCbbc
 170 IF O%=1 PROCbbcsram
 180 IF O%=2 PROCmaster
 190 IF O%=3 PROCsecpro
 ...
 1080 DEF PROCbbcsram
 1090 IF secpro% THEN PROCspoff
 1100 bank%=?eliterom%
 1110 IF bank%=&FF THEN bank%=FNfreebank
 1120 IF bank%=&FF THEN PROCnosram
 1130 PRINT"Loading please wait";
 1140 X%=bank%
 1150 CALLmakerom%
 1160 *RUN INTRO
 1170 ENDPROC
 ....
 2480 DATA "BBC Model B SRAM Elite"
 2490 DATA "Enhanced Version"

Works fine on my B+ 64k (128k upgraded)

He's the disk:

ELITE.SSD

(200 KiB) Downloaded 65 times

[Lardo Boffin]

Can you add a photo of the disc / box?

[rmbrowngr]

Can you add a photo of the disc / box?

It's a disk image I came across and not from physical media.

It either came on the USB stick that came with the Gotek I purchased from RetroClinic, or I downloaded from somewhere (StairWayToHell???). I really don't remember.

[Lardo Boffin]

Ah. I think I got one of those with my gotek from Retroclinic.

[billcarr2005]

Sounds like the " Master 128 & "Executive" 2ndPro Elites " disk image available from Ian Bell's site

http://www.elitehomepage.org/bbc/index.htm

[trixster]

I think the swram version just eliminates disk loading when you go too and from a space station.

[KenLowe]

I tried to run a copy of this image on three different machines last night, but they all hung up at the same point of of launching the ship. I never got to the spinning planet. What confuses me is that these images have worked in the past...

First machine didn't work, so I removed my new IntegraB board, leaving it with only a Retroclinic 1770 controller and Speech upgrade. I then tried to load it from SD Card using a ROM version of MMFS v1.41. It made no difference. It still hung.

On the second machine, I only have the speech upgrade installed. Tried again with MMC / SD Card interface and ROM version of MMFS. Same deal.

On the final machine, I have a Datacentre, so tried importing the image and running from RamFS. Same issue. On this machine I also have a 8271 controller with a GoTek drive. Same issue loading from there (using both DFS 0.98 & DNFS 1.20). Also tried it with MMC / SD Card interface using both ROM version and SWRAM version of MMFS v1.41. Again, same issue. I disabled various ROMs during these tests to keep PAGE as low as possible (usually around &1900). This machine has IntegraB, VideoNuLA, Speech, PiTubeDirect, Econet and BeebSCSI installed. I also tried the SWRAM version on this machine. It hung as it tried to load the SWRAM image (IIRC). Similar to the standard version, the 2nd processor version also hung on this machine when trying to launch the ship. This version of Elite has definitely worked on this machine in the past.

Other than the obvious common speech upgrade (I'll remove that later today and try again), any idea why I might be having this issue? Could this be a timer issue caused by a faulty VIA??? Odd that it would appear on all 3 machines.

[marcusjambler]

The 2nd copro version is working on my beeb thats pretty much the same set-up as yours. (Integra B,speech,vnula,RC 1770,pizero)

The original and SWRAM versions hang as you've described.

Just holler if you want me try stuff out.

Marcus

[rmbrowngr]

I’ve tried both the original and SWRAM versions on my B+ 64k upgraded to 128k e.g. with SWRAM, and both work fine. Haven’t tried yet the version for the copro yet.

[KenLowe]

Thanks for the feedback, guys. I'll dig into this a bit further when I get a chance. I seem to recall having these issues before, but for the life of me, I can't remember what I did to overcome them!

[Kapitány]

Hi there! I'm quite a newbie BBC Micro user. Bought a Beeb B with an SD card adapter that connects to the User port and uses MMFS. I tried 4 different versions of Elite: two was "preinstalled" in the MMB file, original Elite and some Tube version... in case of both, when I would jump to another system, (or maybe earlier at undocking?) "Disk not formatted error" is the result, but I can save my game and load.

I found two other versions, the Elite128 version with the menu at start, and Elite-a. Elite-a did not work, but with Elite128 I can hyperspace jump and play nicely, but when I'd like to save, or view the catalog, nothing happens. Do you have any good Elite version that works perfectly with the aforementioned MMFS and SD card?

Thanks in advance, I can't wait do discover the potention of this fantastic machine!

K

[rmbrowngr]

Hi there! I'm quite a newbie BBC Micro user. Bought a Beeb B with an SD card adapter that connects to the User port and uses MMFS. I tried 4 different versions of Elite: two was "preinstalled" in the MMB file, original Elite and some Tube version... in case of both, when I would jump to another system, (or maybe earlier at undocking?) "Disk not formatted error" is the result, but I can save my game and load.

I found two other versions, the Elite128 version with the menu at start, and Elite-a. Elite-a did not work, but with Elite128 I can hyperspace jump and play nicely, but when I'd like to save, or view the catalog, nothing happens. Do you have any good Elite version that works perfectly with the aforementioned MMFS and SD card?

Thanks in advance, I can't wait do discover the potention of this fantastic machine!

K

You could tried the disk in an earlier post in this thread, or this from StairWayToHell.

Elite.zip

(72.6 KiB) Downloaded 32 times

Or this from the complete bbc micro games archive.

Disc999-EliteA.ssd

(200 KiB) Downloaded 35 times

Or try downloding a version from this site http://www.elitehomepage.org/bbc/index.htm

[rmbrowngr]

And those with money to burn https://www.ebay.co.uk/itm/ACORNSOFT-EL ... Swk7Zcfnn3

[KenLowe]

The 2nd copro version is working on my beeb thats pretty much the same set-up as yours. (Integra B,speech,vnula,RC 1770,pizero)

The original and SWRAM versions hang as you've described.

Just holler if you want me try stuff out.

Marcus

Ok. So, I've made a little bit of progress. On the SWRAM version, it turns out that ELITE was trying to install code into a non existent RAM bank (bank 13 in my case). This was the first empty ROM slot in my system when counting down from the top (bank 15). I installed a RAM module into that bank, and I now get beyond the two dots at the loading screen. I suspect a physical ROM in that slot would have pushed the problem down to the next empty bank. This is probably why I had it working before; I had ROMS / EEPROMs in all slots down to the in built RAM banks 4 thru 7.

It still hangs when trying to launch the ship though! Need to investigate that further now. It did work before.

On the plus side, though, it solved my Exile issue too, which was also hanging at the start when trying to run it in SWRAM mode!

Probably need to add a note into the User Guide to make people aware of this.

[vanekp]

I always though it searches the ROM slots till it find a slot that contains RAM I.E. that it can write to and read from, if I remember well Exile does that as I use to have a watford ROM/RAM board before and had to change the code to get it to recognise the RAM slots on the board. So add that it trys to use a slot that is not A RAM slot in your case.

[KenLowe]

Yes. It's definitely incorrectly identifying an empty bank as SWR. I did start to look into why it was happening, but got a bit confused by the DEF FN code that was being used for this. I'll maybe work on this a bit more tonight.

On another note, I've solved my hanging issue too. Press f0 to start; not space or enter!!!

[vanekp]

does help if you press the right keys

[rmbrowngr]

On another note, I've solved my hanging issue too. Press f0 to start; not space or enter!!!

Not a bad idea to get the function key strip. You can download it from here http://www.8bs.com/othrdnld/manuals/functionkeys.shtml

[1024MAK]

Systems can falsely detect RAM if stray capacitance causes the voltage levels on the data bus to be retained between a write and an immediate read (or sometimes for a slightly longer period) if there is no selected device to drive the data bus.

The software will then think it has detected RAM, as the read back value is equal to the written value.

Note that if bus buffers are used, it’s the inputs on the memory side that cause the problem.

Sometimes pull-up or pull-down resistors can help.

Mark

[KenLowe]

Systems can falsely detect RAM if stray capacitance causes the voltage levels on the data bus to be retained between a write and an immediate read (or sometimes for a slightly longer period) if there is no selected device to drive the data bus.

The software will then think it has detected RAM, as the read back value is equal to the written value.

Note that if bus buffers are used, it’s the inputs on the memory side that cause the problem.

Sometimes pull-up or pull-down resistors can help.

Mark

I suspect this is exactly what's happening. The IntegraB has buffers on both the address (74HCT244) and data (74HCT245) buses with 47k pull down resistor arrays on the memory side.

I'll look into the Elite SWRAM detection code a little bit further this evening to see if I can validate this theory.

[KenLowe]

So, here's the code that's checking for SWR. Addresses &7400..&740F for banks 0..F are all initially set to 0. They are set to &FF if SWR is detected. Empty banks on the IntegraB board are being detected as SWR, and the associated address is being set to &FF. Empty banks on the mainboard are correctly *not* being detected as SWR, and remain set at 0. I'm guessing if I add a small time delay between the store and compare, it would all start working correctly:

Edit: Adding a time delay didn't work! The IntegraB started to correctly report empty slots as not being SWR, but the mainboard empty slot started being reported as SWR!!! So I'm guessing the delay code is also impacting what is on the data bus at the time the empty ROM slot is being 'read'.

I replace the EOR #&01 at &763E with LDA #&AA. Coupled with the time delay, this seemed to give the right answer for all banks, but all a little bit flakey.

[sweh]

That code will probably detect FLASH memory as RAM as well, because FLASH can remember written data for a small while, even if it's not in programming mode.

I had to add this to my code:

Code: Select all

\ FLASH RAM chips aren't sideways RAM; they require a special command
\ sequence to cause them to enter WRITE mode.  But they can "buffer"
\ results, so a write to the chip may be read back again and the written
\ value returned, even though it never made it to the FLASH storage.
\ But if we pause for long enough (200us?) then the cache is cleared
\ and stuff works as normal
\ Through experimentation a loop value of #&50 is too low; #&80 seems to
\ work.  Let's double that.

.delay_swr_write
        JSR SWR_write
        LDX #0
.delaylp
        DEX
        BNE delaylp
        RTS

[KenLowe]

I also thought that might be the case, but I had an EEPROM (28C256) and Flash (29C256) in my board when I was doing this testing, and they passed without any problem.

...but just thinking about it, I had these modules write protected, so that's why they will have worked.

[1024MAK]

The more reliable way round this problem, if, and only if, the data bus that has the empty sockets also has at least one actual ROM/EPROM chip, is to write to the unknown "ROM/RAM" slot, read from a known ROM/EPROM, then try to read from the unknown "ROM/RAM" slot that is being tested. of course, if the sideways ROM/RAM chips are spread across two boards, if the software thinks that RAM has been detected, a second read should be carried out after a suitable delay, as the ROM may not be on the same board as the empty socket that looks initially like RAM...

This may not work with some EEPROMs/Flash ROMs due to the effect that Stephen talked about in the earlier post, these require an even longer delay. However, writing to multiple locations with known values, then reading back those same addresses and comparing may work (I've not tried this) and be quicker.

Mark

[iainfm]

Just came across this while looking for something else!

I bought a copy of Elite around 1986 and it supported a standard B, a B+ with sideways RAM and a Master. I don't recall seeing the same menu that's in this disk image though.

I spent ages trying to get the B+ version to work on our B with a Solidisk 256K (I think) sideways ram extension, but without success. Mainly because I had no idea what I was doing I did get to play the Master version one lunch hour at school though.

Unless my parents have had a tidy out, I should still have the original box and contents somewhere...