Why can't my ESU v4 be reblown?

[mkrob]

It would seem that my ESU v4 sound decoder I wanted get re-blown is obsolete. I emailed a company to ask about getting my chip re-blown with class 22 sounds & they said "The class 22 is a new project and most new projects are on v5 decoders only". So is my v4 decoder of any use apart from a door stop? I was under the impression the main difference between v4 & v5 was the memory capacity, with the ability to have more functions or am I wrong!

Edited September 2, 2020 by mkrob

[Dungrange]

As far as I'm aware, each sound producer customises their sound files to work with one specific decoder, so for those sound providers who use ESU decoders, that will be either v 3.5 , v 4 or v 5.  Presumably all new sound projects will be set up for the new v 5 decoders (as the others are no longer being produced by ESU), but I don't see any reason why you can't get your v 4 decoder re-blown with an older sound project (ie one that was recorded a couple of years ago when the v 4 decoders were being produced by ESU).

[Phil Bullock]

Howes have a Class 22 sound project that will go on a V4 chip. No connection other than satisfied customer....

[mkrob]

20 minutes ago, Richard Croft said:

They can be converted forwards but not backwards, so if you have a file for a v3.5 you can convert it to work on a 4 or even a 5, but you can't go the other way, if it was made for a V5 then thats the only decoder it will work on without starting from scratch, and nobody is going to do that for a one of £20 reblow

So you cant just delete some functions so it will fit on a v4 decoder,like deleting files on a SD card? Very convenient for ESU!

[njee20]

Not really. That’s like saying why can’t i reduce the file size of my MP3 by deleting the guitar parts. 

[mkrob]

After reading up on v5 decoders,it seems that not only do they have more memory,they have better bandwidth so they are up to the standard of hi-fi. So it makes sense that that they are not backward compatible!

[spamcan61]

21 hours ago, njee20 said:

Not really. That’s like saying why can’t i reduce the file size of my MP3 by deleting the guitar parts. 

..but you could reasample your MP3 at a lower bitrate, leaving the music more or less intact but with a smaller filesize. Presumably the same could be done with a loco sound file, although whether there's a business case for doing that is another matter of course.

[SRman]

On 03/09/2020 at 06:33, mkrob said:

It would seem that my ESU v4 sound decoder I wanted get re-blown is obsolete. I emailed a company to ask about getting my chip re-blown with class 22 sounds & they said "The class 22 is a new project and most new projects are on v5 decoders only". So is my v4 decoder of any use apart from a door stop? I was under the impression the main difference between v4 & v5 was the memory capacity, with the ability to have more functions or am I wrong!

In the absence of a suitable file for your v4 (if you don't want the Howes version), you could buy a new v5 with the sound file you want on it, and reblow the v4 for use elsewhere. I have done this on a few occasions, including with v3.5 decoders. There are still a few v3.5 projects available from various suppliers.

[NIK]

9 hours ago, spamcan61 said:

..but you could reasample your MP3 at a lower bitrate, leaving the music more or less intact but with a smaller filesize. Presumably the same could be done with a loco sound file, although whether there's a business case for doing that is another matter of course.

Hi,

 

The sound chip may use a fixed rate of sending digital data to be converted to analogue. If so then the sound can't be compressed to fit. Also MP3 uses a compression algorithm that hides  much the missing information from the listener as possible. It is unlikely that loco sound chips do this.

 

Regards

 

Nick

[spamcan61]

2 hours ago, NIK said:

Hi,

 

The sound chip may use a fixed rate of sending digital data to be converted to analogue. If so then the sound can't be compressed to fit. Also MP3 uses a compression algorithm that hides  much the missing information from the listener as possible. It is unlikely that loco sound chips do this.

 

Regards

 

Nick

I was wondering, for no particular reason, whether DCC sound chips stored uncompressed audio (memory intensive but low processor load on playback) or stored compressed audio that's then decoded on the fly for playback, much less memory but needs more processor power. I'm a Duette user so no skin in this game  but as an electronics designer it did pique my interest.

[Locoman58]

Technically spoken it is no problem to downgrade a V5 project to V4 as long as you can accept the background noise of the 8bit quality.

But it has to be done manually and this takes a lot of time which just doesn't pay.

[Neils WRX]

I think I'd buy the V5 decoder and then use the V4 in a loco that doesn't have sound (which has a V4 project available), or keep it for another model and use it later.

 

I've done this a number of times, guess its maybe easier with European loco's that are all on ESU's website and a Lok Programmer to tweak stuff.

 

Sorry if this isn't much help.

 

Neil

[NIK]

6 hours ago, spamcan61 said:

I was wondering, for no particular reason, whether DCC sound chips stored uncompressed audio (memory intensive but low processor load on playback) or stored compressed audio that's then decoded on the fly for playback, much less memory but needs more processor power. I'm a Duette user so no skin in this game  but as an electronics designer it did pique my interest.

Hi,

 

I'm guessing sound decoder MCUs when playing multiple sounds at the same time have to take data from different parts of an EEPROM, multiply them by the volume for each sound source add them together and multiply them by the master volume . So there may not be time to decompress the data. Then again they may have encrypted the sound files for commercial reasons but the decrypt may be done by MCU internal peripheral hardware and not have too much of a performance hit. I've never put a logic analyser on a sound chip (I assume most if not all use a serial EEPROM for the sound data).

 

Regards

 

Nick