What firmware version of PTN3460 should I choose?

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

What firmware version of PTN3460 should I choose?

Jump to solution
3,708 Views
kasperhansen
Contributor II

A number of different firmware versions (F1 to F6) are available for this chip, but nowhere are the specific details of each version described.

Which version to choose?

Also please send me the accompanying documents AN11128 (Programming Guide), AN11133 (Flash-over-AUX Utility User's Guide), and AN11134 (DPDC Utility User's Guide) as well as the software tool that is mentioned for setting up registers.

Kind regards Kasper

Tags (3)
0 Kudos
1 Solution
2,613 Views
TomasVaverka
NXP TechSupport
NXP TechSupport

Hello Kasper,

All the app notes as well as firmware updates including implemented changes are available on this website. I would recommend using the latest version (F6).

I hope it helps.

Best regards,

Tomas

PS: If my answer helps to solve your question, please mark it as "Correct". Thank you.

View solution in original post

0 Kudos
11 Replies
2,613 Views
brady_gaughan
Contributor I

If anyone has DisplayPort flickering issues, you should try using the DPCD tool. You will probably see symbol errors. DPCD registers provide a lot of information including link settings and match the DisplayPort spec from VESA. And don't use Fast Link Training use Full Link Training. We learned the hard way.

0 Kudos
2,613 Views
amaurel
Contributor I

The link for the firmware and the description off them is no more valid. May you post it again ?

0 Kudos
2,613 Views
PaulDeMetrotion
Senior Contributor I

The link provided on your response is no longer valid. Where can I find this tool?

0 Kudos
2,613 Views
shawnou
Contributor I

Just wondering about the firmware utility from the above link.  

I scan it with Microsoft Security Essentials it sees it as a trojan.

Is this considered normal or is it something I shall worry about?

0 Kudos
2,614 Views
TomasVaverka
NXP TechSupport
NXP TechSupport

Hello Kasper,

All the app notes as well as firmware updates including implemented changes are available on this website. I would recommend using the latest version (F6).

I hope it helps.

Best regards,

Tomas

PS: If my answer helps to solve your question, please mark it as "Correct". Thank you.

0 Kudos
2,613 Views
stefanmattes
Contributor II

We also need actual firmware and tools because of flickering 17.3 inch FHD displaymodule.

PTN3460 we have got with final hardware has an development version on it.

Please can you provide an active link again.

Thanks

Best Regards

Stefan

0 Kudos
2,613 Views
maximilianschin
Contributor I

Hello Stefan,

did the firmware update fix your problem with the flickering display? If so, which firmware version did you update to? We are connecting a FullHD Display to a DP source via a PTN3460BS and have problems with flickering displays too.

Thanks and best regards

Max

0 Kudos
2,613 Views
stefanmattes
Contributor II

Hello Max,

yes it works well, we could not see any flickering. To check this, we took a camera recording changes on the screen.

The old firmware was F2 and we updated to the new F6.

Best Regards

Stefan

2,613 Views
kasperhansen
Contributor II

Hi Stefan. 

I just sent you a PM. 

I have the files, and if you mail me I will send them to you. 

My adress is in the private message. BTW, we had to do some hacking to the Linux core in order to force the link eDP link training to occur. 

I suppose a non-trained eDP link could be the root cause of the image flickering since it could be a boundary condition signal integrity wise. We had no image at all before enforcing the training, but maybe our matching of the differential lines were more 'off' than yours is. 

I can supply some more info on how we solved this. It was our Linux guy that did it, so I have to ask him 

2,613 Views
stefanmattes
Contributor II

Hello Kasper,

sorry for my late answer. Thanks a lot for your help and for the firmware. It works well.

We found version F2 in the chips and we installed version F6.

No flickering again, we are happy with this and our customer also.

Our problem ist solved.

Thanks and best Regards

Stefan

0 Kudos
2,612 Views
kasperhansen
Contributor II

We are using the Apalis TK1 module from Toradex. The following link is a description from Toradex on the issue:

linux-toradex.git - Linux kernel for Apalis and Colibri modules 

Our Linux guy wrote the following when he checked in the altered Linux file thingy:

I have been debugging the driver a bit to see how the link training worked. 
According to the standard, there are 2 parameters to adjust: 
Voltage Swing (values 0.4, 0.6, 0.8, 1.2) and Pre-emphasis (values 0dB, 3.5dB, 6dB, 9.5dB). 
The driver initiates the link training, but it is the sink (i.e. the PTN3460 chip) that 
chooses which parameter values to use.  The link training has 2 phases: Clock recovery and channel equalization. 
The clock recovery would always succeed with the lowest settings, 
but the channel equalization would not. The PTN3460 would then step wise 
instruct the driver to turn up the voltage swing parameter – but the link training 
would not succeed.  After trying all 4 values of the voltage swing, I would expect the PTN3460 chip to 
reset the voltage swing and increase pre-emphasis – but it did not. 
Instead it bailed out and to the driver, the next step would be to decrease the pixel clock. 
The pixel clock could, however, not be lowered because it would then not be able to carry 
the full data rate (on the 2 channels), and hence the whole procedure was aborted.  What I did, was to hack the driver use pre-emphasis 3.5dB and then it succeed 
(with voltage swing 0.4). This seems to work every time and seems to be stable. 
There are 2 things I do not quite like about this solution:
    1) Not all combinations of Voltage Swing and Pre-emphasis are allowed. 
       When increasing Pre-emphasis one step, then one step (the largest) is removed 
       from the allowed Voltage Swing. By just forcing pre-emphasis to 3.5dB 
       (without the chip knowing), we could end up in a situation where the chip requests and 
       illegal voltage swing, and there is no logic in the driver (right now) to catch this.     2) I must admit I don’t quite understand the driver fully. For instance it has internally 
       two arrays of 4 elements – one for Voltage Swing and one for Pre-emphasis. 
       It seems, though, that all values in the arrays are set to the same value. 
       And why would you need an array for *each* parameter? You only need to send one 
       value to the driver!?    Basically, not understanding the driver fully, it makes me wonder whether it is the chip that 
fails to adjust the pre-emphasis, or if it is the driver that fails to recognize instructions 
from the chip to adjust the pre-emphasis.
But right now, I think we are happy that it seems our hardware works :-)
0 Kudos