Improving MP3 Playback Speed - Building an Amiga Sound Card - Part 3
Вставка
- Опубліковано 1 жов 2024
- Join this channel to get access to perks (There are several early access videos) :
/ @gadgetuk164
or you can support the channel via / gadgetuk164
or ko-fi.com/gadg...
Looking back at the Prelude Sound Card and the slow MP3 issue...
Glen (CRG) - • Prelude Zorro 2 Amiga ...
Newer Driver (Library) - www.a1k.org/fo...
Part 1 - • Building an Amiga Soun...
Part 2 - • Fixing the MPEG... Bui...
#commodore #amiga #upgrade
Kavanoz kindly provided a link to order the crystal - www.mouser.com/ProductDetail/ECS/ECS-100AX-147.4?qs=GxOUx7aO6nzPauKUohZfeg%3D%3D&countryCode=US¤cyCode=USD
Note: The updated driver (library) I believe just handles the difference in clock speed when using a 14.318Mhz crystal. There is a revision difference between the MP3 decoder fitted at the start of Part 2, and the F10 revision fitted by the end of Part 2 though. I saw some references to that in the datasheet, and in the A1K thread. I would suggest the F10 revision is better, and I would go with the 14.318Mhz from the start if building and go with the newer driver (library). Personally I am happy with this new crystal and the ~0.21% speed inaccuracy.
Glen (CRG) - ua-cam.com/video/EVzZZAsN4tI/v-deo.html
Newer Driver (Library) - www.a1k.org/forum/index.php?threads/85465/page-18#post-1759074
Part 1 - ua-cam.com/video/hRnVjtIwi1Q/v-deo.html
Part 2 - ua-cam.com/video/X9wJaO2XbPw/v-deo.html
Its a cool card but it is soo long and sparsely populated one would think the daughterboard could be elimenated and the mp3 decoder could live directly on the main PCB. Removes the connector failure point and reduces the height of the card. Its funny the card sag reminds me of the obese multiple card slot GPU's we have today from Nvidia which need their own anti sag brackets. Awesome video. Prelude, Amiblaster, AmigaSID all avaliable today! So much for Paula is all you ever could ask for.
Wasn't aware there are two different "revisions" of the MAS chip. Have some on the way for my card, but see now the source does not specify the exact type other than it is the QFP package. Fingers crossed, I suppose. I also noticed the oscillator did not match the speed specified in the datasheet, so I opted to buy some that were programmed to 14,725 MHz. Hopefully it works in the end!
A programmed XTAL will do the job =D
Finally got the card finished up. Had to get an adapter in order to program the mach-chip. Glad to say it worked straight away. Correct speed and no distortion. Probably won't listen to a lot of MP3s on my 2000 and 4000, but fun project. :-)
this is cool .it is fixed.sounds good
People all over the world 😊problem solving
Well done all of
I saw on the screen it said that WAV was low quality and MP3 was higher quality. Between YT compression and my old and knackered ears, they souded the same. Did you notice a difference, Chris ?
Yes, it's definitely clearer than the wav test!
@@GadgetUK164 Cool.
Great vid, & sound advice.
LOL @ sound advice =D
why are you fighting with Xtals when you can just use a DDS?
Yes, I did look at alternative solutions here! I considered also dividing a faster clock down, but couldnt find anything in the right speed. I considered synthesising my own clock too, with some MCU etc. In respect to DDS and ICs that might be able to do that, I am totally unfamiliar with what's available and what would be easiest to implement.
@@GadgetUK164 I would think something along the lines of the LTC1799 chip may do what you want. Not sure though if the frequency drift of that chip would be a problem or not though. Simon The Magpie uses this chip to modify musical instruments so I assume it is good enough.
To see if the playback speed is still correct I'd recommend creating a 1KHz wav and compressing it to MP3 (I'd even try different bitrates to check it's consistent), then playing back the 1KHz MP3 on the Prelude into either a frequency counter or your scope. I can't say for sure, it could be mind tricks, but the card playback sounded just slightly flat to me...
Oh! You did something similar ;-) I was listening away from the screen and missed that part... DOH!
Actually, does the Prelude use the MP3 daughterboard for WAV playback, or does it simply pass the PCM through to the Crystal DAC? I have a funny feeling your 1KHz test may have only tested the DAC's accuracy and not the MP3 decoder...
That's exactly what I did lol =D
No, the Prelude uses the Mpegit board only for mpeg! The wav is produced by the Crystal IC! And the audio output from the mpeg decoder goes into the DAC on the mpegit. It is analogue by the time it passes out through the Crystal aux port.
@@GadgetUK164 Ah, so the 1KHz test tone was an MP3 and not a WAV? The WinAMP screen was a ruse! ;-)