Internal & External Volume Mixing Levels

Discuss hardware-related topics, such as development cartridges, CopyNES, PowerPak, EPROMs, or whatever.

Moderator: Moderators

Post Reply
Great Hierophant
Posts: 780
Joined: Tue Nov 23, 2004 9:35 pm

Internal & External Volume Mixing Levels

Post by Great Hierophant »

As should be fairly well-known in the community, different revisions of the Famicom and AV Famicom output different volume mixing levels when internal audio is mixed with expansion audio. This is the best video I have found that gives an audio comparison of the widest variety of models :

http://en.nicovideo.geo.jp/?sm16346242

However, if you don't want to suffer through NicoNico's viewing restrictions (account required to watch videos on the site directly, cannot go ahead without a paying membership), I recorded the audio off the video and shared it here with explanatory notes :

http://www.mediafire.com/file/uggbavh67 ... _Levels.7z

Here are a pair of page which provide shots of many of the PCB revisions to the Famicom (mostly taken from other sources) :

http://jpx72web.blogspot.sk/2016/11/fam ... d-new.html
http://offgao.blog112.fc2.com/blog-category-4.html

Ultimately, to my ears there is no appreciable difference until you get to the GPM revisions, where it tends to stay more or less consistent until the Sharp units. The volume level of the internal audio in these units is quieter than the earlier units, but the volume level tends to be mostly consistent for all the later Nintendo manufactured units, including the AV Famicoms. The Sharp units sound muffled and sometimes distorted, but that could be due to the output path after mixing. Even so, the volume levels seem to be more equal on the Sharp units than the later Nintendo units.

So what is the explanation for the drop in internal audio levels in the later units? The mixing circuit seems remarkably stable throughout the Nintendo units, it goes through mixing resistors, an electolytic capacitor (same values appear to be used throughout for the resistors and capacitor) and an inverter and then to the cartridge connector. I used to believe that the revisionless and revision E CPUs on the early Nintendo and Sharp boards output volume at a higher level, but now I am wondering if the extra capacitors on the later Nintendo boards (43 vs 14) may be attenuating the volume signal?
Post Reply