forum.defence-force.org

jdavis6809 wrote: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

The ripple are caused by fast switching, the slower the front are (rising or falling) the less ripple we will have on the line. The microdisc use old TTL and CMOS chip which have "long" switch time, modern component, like the one use in the cumulus have shorted switch time.

This has reminded me of a modification that Dave Wilkin came up with for the Microdisc.
Would this be of any use???

Picture attached

It is only used to have a stronger signal for the 6502 clock.

The same that with Cumulubus or AmpliBus, but with AmpliBus or Cumulubus you don't have to open the Oric. Also, both regenerate other signals more.

Interesting stuff.
I can see I'm going to have to buy an Amplibus. Is anyone building these or supplying kits ?

Well, I'm thinking i must open here, a list of interested people.

Hello,

can any one do a test in particular of working cumulus and none working cumulus

Test 1 with amplibus at oric end and cumulusbus and the other end ( yes requires 1 amplibus and 1 cumulusbus)

cannot do a test with two cumulusbuses because one connector is 34way and the other connector is 40way

try with different length cables say 150mm, 300mm, 450mm

this will test the system with normal 34way ide cable


Test 2. with Cumulusbus at oric end via a 40way ide cable to the cumulus end ( yes requires 1 cumulusbus)


try to use different lengths of 40way ide cable as possible

try to use different lengths of 80way ide cable as possible

try to use different lengths of Round 80way ide cable as possible (twisted pair)


this will test the system with normal 40/80way ide cable


some people may have more than 1 cumulus that could do this test

the only way you can use two cumulusbuses is to make a 34way to 40way adapter

regards

john davis

All looks very good!etui silicone Xperia M4 Aqua coque galaxy S6 protection

Hi guys. Just wanted to share something that was posted in RetroWiki regarding Cumulus and its related issues.

One of the users (ron) had some minor problems with Cumulus, where the screen was filled with artifacts when loading data from disks. After some testing it seemed it was related to the generated MAP signal timings. Silicebit suggested changing the ULA of the Oric with another one ron had as replacement and voila! it started working flawlessly.

The original one (giving problems) was labeled "CDI/8 8302" and the new one (with no problems) was labeled "CDI 8F 8422". So there may be some slight differences between ULAs. If you are experiencing problems with your Cumulus and you have a replacement of your ULA, it is worth a try.

Interesting, I will try with the differents ULA I have, but that show that I was right, the two amplibus in a row are just covering the real problem.

The problem may not directly be the MAP signal, but indirectly from another signal (mainly Phi2) that make the cumulus not delay map correctly.

In fact the way the cumulus CPLD use Phi2 could be one of the problem if it not perfectly clean.

I hope that my idea will correct the problem and I would offer a way for existing Cumuluses to solve this without change on the PCBs


Edit: do you have a link to the retrowiki page about this?

Edit2: Is that normal that I got this from the main page of retrowiki:

Forbidden
You don't have permission to access / on this server.

??

Yes, there was no doubt, that two buffer PCB's don't solve the problem.
My hope was just to be lucky and 2-buffers make my Cumuluses to work - unfortunately no luck .
I have 4-5 ULA's with different labels, later I'll check all of them and will post the results here.

BTW, Godzil, can you share your idea for improved PHI2 usage?

Exactly Godzil. That is what I think too. Silicebit is right and probably if cumulus provided a way to alter the delay of the MAP signal it could solve most of the problems of the device.

But I just wanted to pinpoint that it is a possible hack for those having issues, if they have a replacement for the ULA, that is.

Even we could collect the ids of those ULAs which seem to work. Mine works more or less and it is labeled as: CDI-7C 8316. Remember I have some issues that make me reboot until the disk loads correctly (and then works with no problems), and which are removed after placing a second AmpliBus.

Anyway the link to the forum is: http://www.retrowiki.es/fororw/viewtopi ... &start=100

Chema wrote:Exactly Godzil. That is what I think too. Silicebit is right and probably if cumulus provided a way to alter the delay of the MAP signal it could solve most of the problems of the device.

But I just wanted to pinpoint that it is a possible hack for those having issues, if they have a replacement for the ULA, that is.

Even we could collect the ids of those ULAs which seem to work. Mine works more or less and it is labeled as: CDI-7C 8316. Remember I have some issues that make me reboot until the disk loads correctly (and then works with no problems), and which are removed after placing a second AmpliBus.

Anyway the link to the forum is: http://www.retrowiki.es/fororw/viewtopi ... &start=100

The important part on the ULA marking are

the 1 or two letter after the CDI, which seems to be the revision, and the four number on the next list (8316 on your own) which are basicly YYWW
yy == year (1983 in you case) and WW == week number (16 in your case so)

So your ULA is revision 7C (?) and was build between the 28 of march and the 3 of april of 1983

The "revision" field (7C, 8, 8C) could also be the manufacture that build the chip, it's not know.

@Godzil did not know that information Thanks. My ULA is from 1983, which is correct as my Oric was an Oric-1 once

@iss Please do so and post the results. It would be interesting to see if certain ULAs with specific serial numbers work OK while others don't.

It would be nice if all of you having a Cumulus (working or not) could tell us the ULA version of your Oric and if Cumulus works without problems.

Hi,
I've 6 Atmos but only one matching the Cumulus

The following ULAs do not work with Cumulus

HCS 10017
CDI 9F
8414

HCS 10017 HCS 10017
CDI 7C CDI 7C
8316 8316

HCS 10017
CDI 7C
8315

HCS 10017
CDI 7C
8312

The ULA hereunder works fine with the Cumulus.

HCS 10017
CDI 8F
8401

Hope it will help to fix problems .
Dom.