Cumulus Issues

In this forum you can write about anything that does not fit in other forums.
This includes generic Oric talkings and things that are totaly unrelated but want to share with people here :)
User avatar
Chema
Game master
Posts: 3014
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: Cumulus Issues

Post by Chema »

Hi guys!

I am not sure if I already told you, but my random errors with the Cumulus (minor, but requiring a reset until it worked) disappeared as soon as I pluged also Silicebit's Amplibus (http://forum.defence-force.org/viewtopic.php?f=9&t=1057). Now it works flawlessly with not a single instability.

So adding an Amplibus it is worth a try.
Godzil
Squad Leader
Posts: 774
Joined: Sat May 21, 2011 7:21 pm
Location: Between UK and France
Contact:

Re: Cumulus Issues

Post by Godzil »

The cumulubus should do the same thing... It may mean the cumulubus have to be reworked..

I will look at that
User avatar
iss
Wing Commander
Posts: 1641
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: Cumulus Issues

Post by iss »

wow, Chema, this is interesting - if I understand correctly, you connect first Silicebit's Amplibus to Oric, then you plug Cumulus buffer board and the Cumulus itself ? And this works for you? If so is it possible to get Silicebit's Amplibus schematics?
User avatar
Chema
Game master
Posts: 3014
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: Cumulus Issues

Post by Chema »

Yes, but that worked for me, probabaly because it smooths the signals or delays them a bit more. Not sure.

About the schematics, I think that it is a normal amplibus, and its schematics are availabe (can't remember where but Google for them). Or maybe Silicebit can answer this, or even provide one for you. They were really cheap.
Last edited by Chema on Tue Jan 20, 2015 1:36 pm, edited 1 time in total.
User avatar
iss
Wing Commander
Posts: 1641
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: Cumulus Issues

Post by iss »

Thanks Chema (& Google) :).
Attaching Silicebit's Amplibus schematics here for everyones reference.
I'll try it with my not working cumuluses.
esquema_amplibus.png
User avatar
coco.oric
Squad Leader
Posts: 720
Joined: Tue Aug 11, 2009 9:50 am
Location: North of France
Contact:

Re: Cumulus Issues

Post by coco.oric »

Thanks for the infos.
I'll try the silicebit's amplibus on a couple :
cumulus - oric which is not working (i've my cumulus which aren't working with some oric and working with others)

Didier
coco.oric as DidierV, CEO Member
Historic owner of Oric, Apple II, Atari ST, Amiga
Godzil
Squad Leader
Posts: 774
Joined: Sat May 21, 2011 7:21 pm
Location: Between UK and France
Contact:

Re: Cumulus Issues

Post by Godzil »

"Bad news" the CumuluBus and the Amplibus are.. the same thing (apart from the fact that the cumulubus provide a HE40 header instead of the HE34 with the same pinout as the Oric. So there is little to do on this side.. Adding two set of buffer in a row is not a real solution.
User avatar
iss
Wing Commander
Posts: 1641
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: Cumulus Issues

Post by iss »

Indeed, they are the same and using both makes no sense,
but if it works in some case this could mean that something is wrong with the timings.
User avatar
Chema
Game master
Posts: 3014
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: Cumulus Issues

Post by Chema »

iss wrote:Indeed, they are the same and using both makes no sense,
but if it works in some case this could mean that something is wrong with the timings.
Agreed. It may make no sense, but it works for me, I can assure that.

Probably some additional kind of filtering or signal conditioning or whatever is what my Oric needed!

Basically, if you have an Amplibus at hand and experience issues with Cumulus, it is worth a try...
User avatar
ibisum
Wing Commander
Posts: 1646
Joined: Fri Apr 03, 2009 8:56 am
Location: Vienna, Austria
Contact:

Re: Cumulus Issues

Post by ibisum »

I have two CumuluBus, they both work fine on two Atmos'. Haven't tried with Oric-1, its too delicate an operation, as the Oric-1 I have needs repairs first.
jdavis6809
Pilot Officer
Posts: 80
Joined: Wed Feb 11, 2015 10:39 am
Location: uk

Re: Cumulus Issues

Post by jdavis6809 »

Hello,

I am an electronic engineer of 30+ year (hardware 80% software 20%).

The problem relates to the data bus not been buffered at the oric end.

I cannot see how anyone missed this.

Retromaster stated everything worked with a short cable ( long cable becomes a problem) tries to use PC 80 ide cable to minimise the problem . all ide cables have capacitance an at 1Mhz this can cause reflections, distortion of the signal (twisted pair PC 80 ide cable would have been better , low capacitance).

At the cumulus end everything is buffered and that's ok.

if you look at the the bbc micro 1Mhz bus, what do you see all address, control, data lines are buffered.

if you look at the coco, dragon 32 / 64 the same is true ( seen a coco sd card interface on the internet works the same as cumulus without the display).

if you look a the Cumana drive the control box is close to the oric (I wonder why ?) and a cable from that goes to to the floppy drive. So it is obvious that Cumana had problems so they design it like it is to get round the problem.

Also why on the amplibus (both versions) there is no provision for extra expansion sockets (like version 1)

Also amplibus version 1 had 4 chips (I wonder why ?, see attached) and 3 expansion sockets.

Also it is beyond me that the people who designed the Oric /Atmos, left the 1Mhz bus not buffered ( other than to save money)

Also how did the Oric Microdrive manage without problems (unbuffered at Oric end, buffered at Oric Microdrive end).
Is the ide cable short? If so there's your answer.

There other thing you can do and that's with buffers use termination resistors like they do with ide hddrives.
should not need to do here at 1Mhz I don't think.

They did right when they the designed the microtan 65 system (I had this system when it came out).

Just to note do not have a cumulus but would like one (I have oric 1 48k still looking for atmos 48k)

I done some googling for parts for this project the lcd display is from a nokia 6610 mobile phone, cpld ok, pic ok, rom ok. dbl buffer ok,

So there is no problem to build this

Regards

John Davis
Attachments
1 amplibus-comp.jpg
User avatar
ibisum
Wing Commander
Posts: 1646
Joined: Fri Apr 03, 2009 8:56 am
Location: Vienna, Austria
Contact:

Re: Cumulus Issues

Post by ibisum »

Hey - great to hear from you and have your insight into the situation!

I'd love to hear more about your experience with the Microtan 65. I'd only ever read about it in some scant magazines, and mostly it was just the ad.

I have an excellent-condition Atmos for you if you make me a fair price. I picked up a pallet of them some years ago, and can thus spare one for any interested party, which you seem to be? PM me with an offer and your address if you want to go that route.
jdavis6809
Pilot Officer
Posts: 80
Joined: Wed Feb 11, 2015 10:39 am
Location: uk

Re: Cumulus Issues

Post by jdavis6809 »

Hello,

The t.r.a.n. amplibus v1 the pictureI posted is the same as the other visions ie databus not buffered (4 th chip is a hex inverter tt not inverts or inverts signal 3, ph02 via wire links.)

so I still think the data bus should be buffered at the oric end.

regards

john davis
Godzil
Squad Leader
Posts: 774
Joined: Sat May 21, 2011 7:21 pm
Location: Between UK and France
Contact:

Re: Cumulus Issues

Post by Godzil »

There is no bidir buffer on the data line on the Amplibus. In fact the CumuluBus and the amplibus are exactly the same thing, apart from the connector on the ribbon side.

There are a lots of 1 and Atmos that have problem with the microdisc, and the main culprit is Phi2 which is not buffered on the Oric side, where it should be, as stated by MOS Tech (the signal is comming from some analog part in the 6502 and is not clean.

Buffering the Data bus will not clear the ripples, the only way to clear the ripple is to force the cable impedance by using resistors on both sides like we do on "highspeed" bus like SCSI terminators are, they are to prevent ripple on the bus lines.

It's true that adding a buffer on the data line will make signals a bit cleaner, but it will not solve all the problems, and it will cause one problem which is that we in fact need a tristate buffer, because we can't let it drive the data bus if the external is not used/addressed, and it will not make things easy as we will need something comming from the outside to tell when the external device really want to speak.

The expansion bus is directly connected to the 6502, and could be used/respond to every possible address on the bus, we generally use the page 3 for IO, and the last 16K on the RAM/ROM overlay, but we could in fact disable the whole memory, and could allow the external device to provide the whole 64K of memory. There is no way to add a memory address decoding on the "buffer card" as we don't know how de device will work.

Strictly speaking, it will better/easier to clean the ripple on the lines than buffering the D bus.
jdavis6809
Pilot Officer
Posts: 80
Joined: Wed Feb 11, 2015 10:39 am
Location: uk

Re: Cumulus Issues

Post by jdavis6809 »

Hello,

If what you say is correct

Using amplibus or cumulubus at oric end (where every line is buffered except the databus) it should work

But it does and it does not (also what about the 150ns delay between ph2 and /map)

coco sd interface is similar but is un-buffered because the cartridge is a short distance to the 6809 bus and you can get away with it.

But how the microdisc managed surprises me ( seen a picture so I know what the cable length is)


Regards

john davis
Post Reply