TIPI Usage and Support

[+jedimatt42]

On 9/18/2021 at 6:32 PM, J-Data said:

This may be a dumb question, but what’s the downside of setting the TIPI CRUBASE to >1000?  I had assumed this was the better choice so it wouldn’t conflict with a real disk controller.

There are programs that expect the CALL FILES VDP buffer management routine to work. TIPI doesn't need VDP sector buffers so it only manages that if the crubase is >1100. This provides compatibility with the most programs.

 

The number is small, typically hybrid XB bitmap graphics apps, "The Missing Link" to name one.

 

It also prevents the TIPI generation from creating large, run of the mill XB programs that are too big to fit if a real disk controller is present.

 

 

[J-Data]

On 9/19/2021 at 11:51 PM, jedimatt42 said:

There are programs that expect the CALL FILES VDP buffer management routine to work. TIPI doesn't need VDP sector buffers so it only manages that if the crubase is >1100. This provides compatibility with the most programs.

Understood, I'll update the TIPI/32K combo Wiki and start making mine with the first jumper populated. 

[GDMike]

Can I access any of the TIpi troubleshooting logs from the TIpi (TI basic) menu? 

And is their any way to look at TIpi logs without removing the raspberry SD card? 

I have not seen how to get to logs on the forum.

 

Edited October 5, 2021 by GDMike

[+arcadeshopper]

Can I access any of the TIpi troubleshooting logs from the TIpi (TI basic) menu? 
And is their any way to look at TIpi logs without removing the raspberry SD card? 
I have not seen how to get to logs on the forum.
 

Easiest way is the web browser
Tipi:9900

Sent from my LM-V600 using Tapatalk

[GDMike]

8 minutes ago, arcadeshopper said:

Easiest way is the web browser
Tipi:9900

Sent from my LM-V600 using Tapatalk
 

Thank you

[+9640News]

I telnet into my PI and navigate to the logs file from my Windows PC.  Then, with either the tipi.log file or the daemon.log file (I think those names are correct), I can click on the file and open it in notepad.  There are some other log files as well, but I think those are the only two you might really need to review in most cases.

 

The tipi.log file contains a lot of useful information on the blocks/records and file formats  as things are opened, read/written, and closed.  Matt did exceptionally well with his reporting in the log file.

 

On the TI for file saving, if the file open/read/write/close works on regular devices, highly unlikely you will encounter a problem with file i/o.  I do know during the Geneve development cycle with the tipi, @InsaneMultitasker and myself caught a couple of very unique cases of issues in the file i/o Matt resolved as we pushed the TIPI further than what previous developers had done with the 4A.

 

Now, when you start doing socket access testing for characters/strings, now, you will get some very long log files as each time you scan for a character/string, you get a line entry in the log file.  Those files grow very large very quickly if your code is testing the socket multiple times per second.

 

Beery

 

[GDMike]

41 minutes ago, 9640News said:

I telnet into my PI and navigate to the logs file from my Windows PC.  Then, with either the tipi.log file or the daemon.log file (I think those names are correct), I can click on the file and open it in notepad.  There are some other log files as well, but I think those are the only two you might really need to review in most cases.

 

The tipi.log file contains a lot of useful information on the blocks/records and file formats  as things are opened, read/written, and closed.  Matt did exceptionally well with his reporting in the log file.

 

On the TI for file saving, if the file open/read/write/close works on regular devices, highly unlikely you will encounter a problem with file i/o.  I do know during the Geneve development cycle with the tipi, @InsaneMultitasker and myself caught a couple of very unique cases of issues in the file i/o Matt resolved as we pushed the TIPI further than what previous developers had done with the 4A.

 

Now, when you start doing socket access testing for characters/strings, now, you will get some very long log files as each time you scan for a character/string, you get a line entry in the log file.  Those files grow very large very quickly if your code is testing the socket multiple times per second.

 

Beery

 

I'm just trying to determine where I'm failing in my Foxit program at which time I'm creating a file and I error,hang in DSRLNK on TIpi write to DSK4.TESTDB on real hardware with no disk controller card. BUT DOES work in my SNP program.

Foxit has no error  using classic 99 using and a FIAD setting.

My log from TIpi is over on Foxit post 

Edited October 5, 2021 by GDMike

[+jedimatt42]

4 hours ago, GDMike said:

Can I access any of the TIpi troubleshooting logs from the TIpi (TI basic) menu? 

And is their any way to look at TIpi logs without removing the raspberry SD card? 

I have not seen how to get to logs on the forum.

 

That is because the documentation for TIPI is not on the forum, it is here on Github, as described in POST #1: https://github.com/jedimatt42/tipi/wiki

 

If you remove the SD card, you will not find the logs on it... the logs are setup to use a ramdisk so that your SDcard doesn't wear out. 

 

specifically regarding logs: https://github.com/jedimatt42/tipi/wiki/Troubleshooting

[+Vorticon]

A couple of observations on file functions with TIPI and XB.

• The defaults do not seem to be implemented. For example, OPEN #1:"TIPI.TEST", INPUT does not default to DV80,SEQUENTIAL. These have to be specified.
• When opening an existing TIPI file for OUTPUT, it is not overwritten by the new data, again not the expected behavior. EDIT: This actually happens, but only after the file is explicitly closed (normally XB will automatically close all open files when an abnormal termination is encountered, but this is not happening with TIPI files for some reason).

 

[GDMike]

Just a common sense note for the time.

 

For the first time, the price of a Raspberry Pi single-board computer is increasing and we have the global supply chain shortages to thank for it.

 

Eben Upton, chief executive of Raspberry Pi Trading, made the announcement today, but also made it clear the price increase is only temporary and only impacts one model of the Raspberry Pi 

 

In February last year, the price of the Raspberry Pi 4 2GB dropped from $45 to $35 "permanently" and the 1GB model was discontinued. The 2GB model is reverting back to $45 and the 1GB model is making a comeback for industrial customers at its previous $35 price. The reason for this is one of supply chain challenges, with Upton confirming they will only manage to match 2020's shipments of around seven million units this year. The main shortages have been for the Raspberry Pi Zero and Raspberry Pi 4 2GB.4. time.

Edited October 20, 2021 by GDMike

[+jedimatt42]

8 hours ago, Vorticon said:

A couple of observations on file functions with TIPI and XB.

• The defaults do not seem to be implemented. For example, OPEN #1:"TIPI.TEST", INPUT does not default to DV80,SEQUENTIAL. These have to be specified.
• When opening an existing TIPI file for OUTPUT, it is not overwritten by the new data, again not the expected behavior. EDIT: This actually happens, but only after the file is explicitly closed (normally XB will automatically close all open files when an abnormal termination is encountered, but this is not happening with TIPI files for some reason).

 

TIPI only writes on close or delete for record based access... This is on purpose. I am not inclined to change that unless it is a real problem...

 

As for the defaults.. I am surprised. I would have thought BASIC was responsible for creating a fully intentional PAB.  I'll take a look at this.

 

Ah, "Functional Specification for the 99/4 Disk Peripheral" section 4.1.4 backs you up.

[+Vorticon]

33 minutes ago, jedimatt42 said:

TIPI only writes on close or delete for record based access... This is on purpose. I am not inclined to change that unless it is a real problem...

It's not a problem as long as the user ensures a CLOSE statement made prior to exiting the program. Common sense but I have forgotten to do that on many occasions but XB took care of that for me anyway...

[+jedimatt42]

I am aware there is a PI Zero 2W available now. It should work just fine. It uses the same SoC as the PI 3, so I am confident the existing TIPI sd image will be compatible. I do have one on order, so I can upgrade my status from confident to verified.

 

The previous PI Zero W will remain supported.

[jwild]

My Tipi (PEB) hasn't worked for some time now, but today I thought I'd try to fix it.

 

Problem was that I would type CALL TIPI at the basic prompt and it would just hang.   

 

Today I connected a monitor and keyboard to the rpi and could see it running and log into it.   I still got the hung blue screen.

 

I decided to reflash it using the latest image "tipi-sdimage-buster-2.17.zip" - first formatting with SD Card Formatter and then writing the image with Win32DiskImager.  

 

Result: it still hangs exactly the same way.   I do see the red light come on on the PEB card as soon as i hit ENTER after typing CALL TIPI.

 

I've moved the PEB card to different slots - same result

I've removed all other cards from the PEB - same result.

I've tried a different TI-99 - same result.

 

Any suggestions on what to try next?

 

Thank you...

 

[+arcadeshopper]

My Tipi (PEB) hasn't worked for some time now, but today I thought I'd try to fix it.
 
Problem was that I would type CALL TIPI at the basic prompt and it would just hang.   
 
Today I connected a monitor and keyboard to the rpi and could see it running and log into it.   I still got the hung blue screen.
 
I decided to reflash it using the latest image "tipi-sdimage-buster-2.17.zip" - first formatting with SD Card Formatter and then writing the image with Win32DiskImager.  
 
Result: it still hangs exactly the same way.   I do see the red light come on on the PEB card as soon as i hit ENTER after typing CALL TIPI.
 
I've moved the PEB card to different slots - same result
I've removed all other cards from the PEB - same result.
I've tried a different TI-99 - same result.
 
Any suggestions on what to try next?
 
Thank you...
 

Please photo your cable the light hanging is usually bad comm

Sent from my LM-V600 using Tapatalk

[jwild]

It did work a few months ago.

I could ohm this cable out.. haven’t done that yet

I do feel a slight crease on three of the wires in the cable about at the midpoint…. But no obvious breaks.


Sent from my iPhone using Tapatalk

[+9640News]

Is the screw you are using to mount the PI to the TIPI causing that end of the cable on the PI to be lifted up so that the cable is not fully connected?  Looks like from the photo there may be a slight tilt on that cable.

 

Beery

[+arcadeshopper]

9 minutes ago, jwild said:

It did work a few months ago.

I could ohm this cable out.. haven’t done that yet

I do feel a slight crease on three of the wires in the cable about at the midpoint…. But no obvious breaks.


Sent from my iPhone using Tapatalk

 

Likely the cable is fine.. on the PI hooked up to a screen are you getting anything?  It should boot up to a linux login 

 

 

[jwild]

Thanks Beery.   I see what you are saying.   I can feel the connector has a lot of travel in the pins and sockets and don't think the nut in the picture is in the way.

 

I did ohm out the cable and everything checks out.   I didn't flex it much while testing continuity, but everything checked out. 

 

put it back together and same result....maybe I should order another cable.

 

I do get something on the screen attached to the PI.  i see the 4 raspberries, then the linux text as it boots up, then a login prompt.

[jwild]

i also have a sidecar TIPI somewhere around here I could test

Edited October 29, 2021 by jwild

[curtludwig]

30 minutes ago, jwild said:

My Tipi (PEB) hasn't worked for some time now, but today I thought I'd try to fix it.

 

Problem was that I would type CALL TIPI at the basic prompt and it would just hang.   

 

Today I connected a monitor and keyboard to the rpi and could see it running and log into it.   I still got the hung blue screen.

 

I decided to reflash it using the latest image "tipi-sdimage-buster-2.17.zip" - first formatting with SD Card Formatter and then writing the image with Win32DiskImager.  

 

Result: it still hangs exactly the same way.   I do see the red light come on on the PEB card as soon as i hit ENTER after typing CALL TIPI.

 

I've moved the PEB card to different slots - same result

I've removed all other cards from the PEB - same result.

I've tried a different TI-99 - same result.

 

Any suggestions on what to try next?

 

Thank you...

 

Just over a year ago my system had the same problem. Turned out to be problems with the RAM expander which had to be repaired...

[+arcadeshopper]

i also have a sidecar TIPI somewhere around here I could test

Try that and or another cable.


Sent from my LM-V600 using Tapatalk

[chuckcaley]

On 10/29/2021 at 2:54 PM, jedimatt42 said:

I am aware there is a PI Zero 2W available now. It should work just fine. It uses the same SoC as the PI 3, so I am confident the existing TIPI sd image will be compatible. I do have one on order, so I can upgrade my status from confident to verified.

 

The previous PI Zero W will remain supported.

I can confirm that the PI Zero 2W does work with the TIPI.

[+jedimatt42]

6 hours ago, chuckcaley said:

I can confirm that the PI Zero 2W does work with the TIPI.

https://github.com/jedimatt42/tipi/wiki/Which-PI <- the current list I can support.

 

It is less about what does work, and more about what I can debug on and therefore actually support.

[+jedimatt42]

I have added it ( PI Zero 2W ) to the supported list. 

 

I have tested with the year old version of Raspbian that is in the TIPI 2.17 image. Upgrade to TIPI 2.25 took 20 minutes, but mostly because there are a lot of upstream updates in the base OS.