Midi to USB Please help !!!

just to give a little update,

My friend and me have started to test the circuit on his computer with an professionnal audio card and professionnal software (Sonar 4) and it seems to be detected same way that it was with my comp (http://www.midibox.org/forum/index.php?topic=8817.msg62252#msg62252) with the diff that it appears in Sonar 4. During we were starting to test on his computer, he had a crash and memory failure (socket itself, not the memory !!) and we had to stop. So up to now I can’t confirm it works, but with the brief tests we took, it seems to be in the good way.

I must say that even if I install Sonar 4 on my comp, the Midi card doesn’t appear for me… that’s wreid…

Can someone explain me how go get the pic revision version (to know right know if we are subject to the silicon bug or not) ?

Is there anyone who would test the circuit in // with us to concure on inconcure ?

Thank you

I can offer a bit of information. I wrote a USB driver for class compatible interfaces, though not for Windows.

To qualify as “Class Compatible”, and be recognized by the built-in USB MIDI drivers, a device must identify itself as

CLASS 0x01 (meaning Audio) and

SUBCLASS 0x03 (meaning MIDISTREAMING)

Unless both of these match, it will not be handled with the builtin USB MIDI device driver.

I don’t know much about windows, I avoid it as much as i can, but any decent USB sniffer should be able to give you the class and subclass of any device.

Good Luck,

LyleHaze

thank you for this information. That part seems to be ok in the code… at least for my poor knowledge

I must say that even if I install Sonar 4 on my comp, the Midi card doesn’t appear for me… that’s wreid…

Can someone explain me how go get the pic revision version (to know right know if we are subject to the silicon bug or not) ?

The problem here is with Windows or maybe the driver on your PC… The visibility of the port to sonar, is secondary to the port appearing in windows, which it doesn’t on your PC. Sonar, or MIOS studio, both get a list of available midi ports from windows. If it works OK on your friend’s PC, it’s probably not the midibox which is at fault.

The EUSART bug does not effect the appearance of the device in windows either, only the data that is sent later on. There are no working revisions of the PIC.

I think you’ll find that formatting and reinstalling windows will sort this out.

The problem here is with Windows or maybe the driver on your PC… The visibility of the port to sonar, is secondary to the port appearing in windows, which it doesn’t on your PC. Sonar, or MIOS studio, both get a list of available midi ports from windows. If it works OK on your friend’s PC, it’s probably not the midibox which is at fault.

The EUSART bug does not effect the appearance of the device in windows either, only the data that is sent later on. There are no working revisions of the PIC.

I think you’ll find that formatting and reinstalling windows will sort this out.

maybe you are right for the reinstallation. But, for the very little time we had on my friend’s comp, it seems it was beeing display exactly the same way in his devices that it was on mine… that’s why it’s wreid… According to this post, there is at least one working version of the device (see post # 17) http://forum.microchip.com/tm.aspx?m=85120#

But, for the very little time we had on my friend’s comp, it seems it was beeing display exactly the same way in his devices that it was on mine…

In this post, in the second image down, there was no device in the list. You sure that was the same on his PC? Cause if Windows doesn’t show it there, but Sonar sees it, that’s unusual… IE theoretically impossible, because that window gets the list by using the same MMAPI call as sonar…

According to this post, there is at least one working version of the device (see post # 17) http://forum.microchip.com/tm.aspx?m=85120#

I thnk TK has since retracted that :frowning:

In this post, in the second image down, there was no device in the list. You sure that was the same on his PC? Cause if Windows doesn’t show it there, but Sonar sees it, that’s unusual… IE theoretically impossible, because that window gets the list by using the same MMAPI call as sonar…

I thnk TK has since retracted that :frowning:

well those are took on my comp. As I said it doesn’t work for now on my comp. We had not the time to get printscreens on my friend’s comp because he had a severe bug that has corrupted his windows and his motherboard has memory read errors (not the memory itself, but slots !) and we unable to precess more test with his computer (that have Sonar, Professionnal audio soundcard etc.) at the same time we were testing… That’s an illusrated example of the Murphy Law !!!

As long we cannot test, it cannot go further. My friend will probably need to get a new comp (or at least new MB and maybe other parts…). So by the time if anyone would have get more info ?

By the time, I have contacted Microship to get some help here is their response :

  1. Once when you receive the PIC’s you can find out the revision of the PIC Devices.

  2. Please note that you will receive the samples as per the lot available with Microchip.

  3. As per the revision you have the errata which gives a work around for the EUSART.

  4. If you have received Rev A3 then issue has the workaround is as follows:

  Module: EUSART

When performing back-to-back transmission in 9-bit mode (TX9D bit in the TXSTA register is

set), an ongoing transmission’s timing can be corrupted if the TX9D bit (for the next transmission)

is not written immediately following the setting of TXIF. This is because any write to the

TXSTA register results in a reset of the Baud Rate Generator which will effect any ongoing

transmission.

  Work around

Load TX9D just after TXIF is set, either by polling TXIF or by writing TX9D at the beginning of the Interrupt Service Routine, or only write to TX9D when a transmission is not in progress

(TRMT = 1).

Date Codes that pertain to this issue: All engineering and production devices.

For further information please refer to the weblink given below:

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1999&ty=&dty=&section=&NextRow=&ssUserText=PIC18F2455

Hope this helps

Revert back in case of further queries.

Best Regards

If you click on the weblink, you can see that is seems to be an version B4 of the pic and in the errata, there is nothing about the EUSTART bug so it MAY work for the second part… as long the first (detection) is solved…

By the time, I have contacted Microship to get some help here is their response :

Wrong bug, ours is another one…

Wrong bug, ours is another one…

could you be more specific please ?

Re: PIC18F4620: new silicon revision B4

Use the search, luke. Click SEARCH, Search for : EUSART by user:  TK

Your bug is about 9bit mode, that’s not us.

well … sorry. as I said before i’m not a Pic programmer and I want to build the whole thing with the best of my knowledge and I did’t get it about the bug… anyway, by the time I wrote down more explanations for the customer service of Microchip and there is their answer :

You will have to contact our local office.

Initiate your process with our local sales office.

We do not have any documentation for the USART issues.

Our local sales office nearby your area is as given below:

Sales Office

Microchip Technology Inc 

6285 Northam Drive

Suite 108

Mississauga,ON L4V 1X5

Canada

Phone: 905-673-0699

Fax: 905-673-6509

Hope this helps.

Revert back in case of further queries.

Best Regards

personnaly, at this step I can’t go further as I cannnot explain with technical terms to them the problem. I will probably just abandon the idea like others did :frowning:

well … sorry. … I did’t get it about the bug…

Hah, no need to apologise, that makes both of us! That’s why I suggested the search for TK’s posts - I’m not much of an nameity on these things.

personnaly, at this step I can’t go further as I cannnot explain with technical terms to them the problem. I will probably just abandon the idea like others did :frowning:

Yeh unfortunately MC have been less than helpful on this one :frowning: Now you know why everyone else has given up :frowning: