Log inRegister an accountBrowse CSDbHelp & documentationFacts & StatisticsThe forumsAvailable RSS-feeds on CSDbSupport CSDb Commodore 64 Scene Database
You are not logged in 
CSDb User Forums


Forums > CSDb Entries > Release id #155140 : VICE 3.1
2017-04-08 20:14
LMan

Registered: Jun 2010
Posts: 40
Release id #155140 : VICE 3.1

Hey,

would someone please tell me what ever CSDb rule I've broken this time, so that my comment was removed from the Vice 3.1 release?

Thank you
Markus
2017-04-08 20:36
Groepaz

Registered: Dec 2001
Posts: 8024
and mine too!
2017-04-08 20:45
LMan

Registered: Jun 2010
Posts: 40
Ah Groepaz I'm afraid I didn't read your reply before it was removed. Would you kindly repeat it here?
2017-04-08 23:59
CreaMD

Registered: Dec 2001
Posts: 2450
I guess someone wanted to move debate from comments to forum?

Your comments

User Comment
Submitted by Groepaz on 7 April 2017
resid-FP was removed because, quoting alankila, "new resid is better in about any imaginable way". it wont come back, the idea is to improve the regular resid obviously.

User Comment
Submitted by LMan on 7 April 2017
Hey and huge thanks for keeping the best c64 emu alive!!

I do have an issue with the new filter emulation though. While the ReSID FP modes were super accurate to a point where I could not tell any difference to the real 8580, the filters in 3.1 sound drastically different. Are there any plans to reintroduce ReSID FP into Vice?
2017-04-09 01:21
Karmic

Registered: Apr 2015
Posts: 33
Quoting LMan
I do have an issue with the new filter emulation though.


Not really. Go in the SID settings and adjust the Filter bias until it fits your preference.

The only thing I'd really suggest for VICE team, is that for such a large range of values, there should probably be a slider that can be adjusted while you listen to the output.
2017-04-09 01:31
LMan

Registered: Jun 2010
Posts: 40
If you say so...

I've adjusted the bias to -2500 which estimates what my real sid sounds. Still the FP version sounds a lot warmer and closer to the real machine recording.

I also liked the idea of having SID chip profiles matching the filter curves of their real counterparts.

So maybe technically the new ReSID might be better, but for now the old model works better for me. However this is of course a little subjective, so I'd like to know how other musicians take on this.
2017-04-09 15:32
mankeli

Registered: Oct 2010
Posts: 42
I wonder if resid-fp style chip profiles can be done with current resid? I used to use -fp as well, because somehow it felt as more accurate presentation of the sound. Might be that multiple profiles (and the fact that you could easily select a profile that sounded close to "your own sid") just gave an illusion of accuracy.

My understanding is that the mainline implementation is more technically correct. But as Alankila's implementation was apparently tuned by hand, it might have some musical quality that the mainline doesn't have.
2017-04-09 16:50
Groepaz

Registered: Dec 2001
Posts: 8024
Quote:
I wonder if resid-fp style chip profiles can be done with current resid?

no, not really. you could try to match the filter params, however it works entirely different and the implementation might change over time so absolute values are a bit meaningless.
2017-04-11 14:21
Slajerek

Registered: May 2015
Posts: 17
Isn't it that Resid-FP as name states is using floating point internally, and Resid is using short integers? Although I'm not very familiar with recent changes to Resid, but by quickly scoring the sources it seems the Resid's mixer is still using integers, and Resid-FP's mixer is using floats. This is subtle difference, but I believe some ears can catch it...
2017-05-12 15:34
dano

Registered: Jul 2004
Posts: 99
As for the Sound-Discussion i talked with Toggle about the Sounds in Reluge and in general. With the latest WinVice3.1 i could reproduce wrong sounds in the Reluge Soundtrack. Which is strange as for the development phase over older versions of WinVice seemed to replay correctly - even with FastSid.

As for the latest versions i can't use FastSid anymore as the sounds are totally broken now. And even ReSid seems to not reply 100% correctly (compared to before).

So i'm pondering to go back to 2.4 as the sound replay was just better there. Even taking in account of loosing latest changes and improvements.

Still i find it strange that something that might be "more accurate" sounds just not "more accurate" compared to real hardware..?

Something seemed to have happened there, and ot my knowledge it's not just me thinking that. Yet i am not skilled enough to pinpoint reasons or possible changes there.
2017-05-12 16:31
Groepaz

Registered: Dec 2001
Posts: 8024
if you want to help to pinpoint bugs - small "tunes" that play as little as possible and then sound wrong will help a lot. perhaps.

"As for the latest versions i can't use FastSid anymore as the sounds are totally broken now."

and these are the comments that make me very suspicious and dont convice me to listen to vague "its broken now!!!!!" kind of reports - there were no changes to fastsid since forever =)
RefreshSubscribe to this thread:

You need to be logged in to post in the forum.

Search the forum:
Search   for   in  
All times are CET.
Search CSDb
Advanced
Users Online
Moloch/TRIAD
hedning/G★P
The Git/Silicon Limi..
Guests online: 60
Top Demos
1 Uncensored  (9.7)
2 Edge of Disgrace  (9.7)
3 The Shores of Reflec..  (9.6)
4 Coma Light 13  (9.6)
5 Lunatico  (9.6)
6 Comaland 100%  (9.6)
7 Incoherent Nightmare  (9.5)
8 Wonderland XII  (9.5)
9 Comaland  (9.5)
10 Wonderland XIII  (9.5)
Top onefile Demos
1 Dawnfall V1.1  (9.5)
2 Daah, Those Acid Pil..  (9.5)
3 SidRok  (9.5)
4 Treu Love [reu]  (9.4)
5 Tunnel Vision  (9.3)
6 Dawnfall  (9.3)
7 One-Der  (9.2)
8 Hardware Accelerated..  (9.2)
9 Globe 2016 [reu]  (9.2)
10 Safe VSP  (9.1)
Top Groups
1 Booze Design  (9.4)
2 Oxyron  (9.4)
3 Censor Design  (9.3)
4 Crest  (9.3)
5 Camelot  (9.2)
Top Coders
1 S.E.S.  (9.9)
2 Wotnau  (9.8)
3 Axis  (9.8)
4 Crossbow  (9.8)
5 Graham  (9.8)

Home - Disclaimer
Copyright © No Name 2001-2017
Page generated in: 0.416 sec.