MiSTer FPGA Forum

Hi,

on the C64 game "Myth-History in the making", the sample from the the women in the beginning "Welcome to Myth" sounds far too quiet to me? Any opinions?

Thanks.

Switch to SID 8580 with Digifix on, then it's louder.

On actual hardware, wouldn't it be the other way 'round?

Myth is from 1989. I would imagine that they did the SID samples with the newer 8580 in mind.

Retro-Nerd wrote: ↑Sat Jul 27, 2024 7:54 pm

Myth is from 1989. I would imagine that they did the SID samples with the newer 8580 in mind.

No the samples are 6581 based, i cant think of any games with 8580 samples from back in the day.

Your options were put up with very very quiet samples or perform the Digifix with a resistor in the SID socket

On the MiSTer core use either 6581 SID or the 8580 with Digifix enabled

I don't remember that the samples in Impossible Mission were very quiet back in the days, or in Space Taxi. My C64 was from '82. So is there something special about the samples in Myth? And what is the deal with this Digifix?

HerrBerzerk wrote: ↑Tue Jul 30, 2024 9:18 am

I don't remember that the samples in Impossible Mission were very quiet back in the days, or in Space Taxi. My C64 was from '82. So is there something special about the samples in Myth? And what is the deal with this Digifix?

Your C64 from 1982 would have had a 6581 SID

The 8580 SID which was used the C64c was a fixed version of the 6581 but the fix affected how it played samples designed for the 6581. They play but at greatly reduces volume.

The 8580 is really the better chip and has since become the standard for the C64 demo scene etc

The Digifix was just a resistor added to certain legs of the 8580 which made 6581 samples louder again, it also I creased noise though

The 8580 is better in so far as stability and filter predictability, which means it's theoretically superior.

But in practice it doesn't sound quite as potentially fat and saturated as a good 6581 with an optimal filter. Unfortunately the filters (and as important the two related caps at position C10+C11) are incredibly fickle and vary greatly between not only filter revisions, but rather individual production batches.

This has little relevance in the (modern) demoscene where people specifically target the 8580, but songs made for the 6581, if you use one with optimal filter, will sound significantly better and more historically correct of course.

The early long motherboard versions of the C64C actually use 6581.

The 8580 (or rather 6582/6582A/8580 R5) was used in the 64C/C64-II (late, short motherboard version), C64G, C64 (Aldi), C64GS, and C128DCR (CR =cost reduced).

I've not checked what could be the issue with Myth, but maybe I'll check out how it sounds on some of my hardware C64s with a variety of chips and compare it to various MiSTer settings.

lagerfeldt wrote: ↑Tue Jul 30, 2024 11:46 am

The 8580 is better in so far as stability and filter predictability, which means it's theoretically superior.

But in practice it doesn't sound quite as potentially fat and saturated as a good 6581 with an optimal filter. Unfortunately the filters (and as important the two related caps at position C10+C11) are incredibly fickle and vary greatly between not only filter revisions, but rather individual production batches.

This has little relevance in the (modern) demoscene where people specifically target the 8580, but songs made for the 6581, if you use one with optimal filter, will sound significantly better and more historically correct of course.

The early long motherboard versions of the C64C actually use 6581.

The 8580 (or rather 6582/6582A/8580 R5) was used in the 64C/C64-II (late, short motherboard version), C64G, C64 (Aldi), C64GS, and C128DCR (CR =cost reduced).

I've not checked what could be the issue with Myth, but maybe I'll check out how it sounds on some of my hardware C64s with a variety of chips and compare it to various MiSTer settings.

The 8580 can do more than the 6581 which is why it has become the norm now, all the chips sound the same too unlike the 6581 and why we see 8580 only releases now. It won't work on the 6581

Yes some early Cs had the 6581

There is no issue with Myth, the "Welcome to Myth" sample is just 6581 optimised, so you can't really hear on the 8580 unless you enable DigiFix

I suggested to Sorg a while ago to introduce the switchable digifix in the core as it was causes issues in some scene demos by amplifying noise. Glad it made it to Main

Yes, indeed.

My statement was more around loudness - in the real world, as I remember it, the 6581's 'digital' output was significantly louder than the 8580 with digifix, where the core seems to behave in the opposite manner.

rhester72 wrote: ↑Tue Jul 30, 2024 1:14 pm

My statement was more around loudness - in the real world, as I remember it, the 6581's 'digital' output was significantly louder than the 8580 with digifix, where the core seems to behave in the opposite manner.

The 8580 with Digifix sounds fine to me but we have the option of 6581 anyway

rhester72 wrote: ↑Tue Jul 30, 2024 1:14 pm

My statement was more around loudness - in the real world, as I remember it, the 6581's 'digital' output was significantly louder than the 8580 with digifix, where the core seems to behave in the opposite manner.

That is my impression too… I have no proof but shouldn‘t it behave in the opposite manner…? Maybe Lagerfeld can really do a test.

lagerfeldt wrote: ↑Tue Jul 30, 2024 11:46 am

But in practice it doesn't sound quite as potentially fat and saturated as a good 6581 with an optimal filter. Unfortunately the filters (and as important the two related caps at position C10+C11) are incredibly fickle and vary greatly between not only filter revisions, but rather individual production batches.

I think, the title music of „Flimbo‘s Quest“ sound much better on the 6581

Really there is no rule to how loud the digifix should be as its just a hack anyway. The amount of gain was done by Sorg