http://xhip.cjb.net/xhip/releases/v0/b6 ... 6.12.5.dll
the sustain thing and waveshaper filter should be fixed now. the waveshaper filter bug is really confusing, it's a compiler bug. for now i've just replaced the inlined functions with function calls for the two places where the bug occurs. a few cycles slower, but no big deal when we're talking few vs. 10000.
XHip--Please finish your synth!!
-
- KVRAF
- 4669 posts since 26 Sep, 2005 from U.A.E
YAY!!!
-
- KVRAF
- 12615 posts since 7 Dec, 2004
btw, a 3ghz pentium d is tehnsane. i can still kill it on only 16 voices of xhip if i max out all the features. (well, i can kill half of it, dual cpu right..)
actually i'm very unhappy with this new system. when i get up to make my tea during xhip compile, by the time i've stood completely the compile is already finished
actually i'm very unhappy with this new system. when i get up to make my tea during xhip compile, by the time i've stood completely the compile is already finished
-
- KVRAF
- 12615 posts since 7 Dec, 2004
-
- jaaathmaster
- 2690 posts since 1 Jun, 2001 from Marlow, S. Bucks, UK
Awesome!aciddose wrote:http://xhip.cjb.net/temp/public/oomptyloopmpty1.mp3
Music with dinner is an insult both to the cook and the violinist.
-
- KVRian
- 889 posts since 10 Jul, 2004
i started makin some music again....and xhip is rockin 
stable as a rock.....and the cpu usage is still very low.
the cpu other synths use is disgustin.....and its not like they sound any better.
I like the hard mode in the filter trigger.....gives a nice knock.
demos comin soon
stable as a rock.....and the cpu usage is still very low.the cpu other synths use is disgustin.....and its not like they sound any better.
I like the hard mode in the filter trigger.....gives a nice knock.
demos comin soon
-
- KVRAF
- 12615 posts since 7 Dec, 2004
i'm actually amazed. i spent all that damn time designing and building analog gear in order to get some kind of "analog sound" while i already knew this was pretty much an imaginary thing.
(no such thing as "analog sound".. it is just about quality. at this point xhip is above that level of quality in many respects, although has a way to go in others)
so now as the mp3 i posted above shows... it sounds pretty much exactly like the filter i built by a simple adjustment of the filter mod and waveshaper.. nothing really special either.
i do kind of miss some of the effects specific to a series-lossy-integrator filter (xhip has a series integrator, aka multimode) which are impossible with xhip. then again, xhip can do things impossible with a series-lossy-integrator.. so..
by the way, if you see http://xhip.cjb.net/xhip/todo.cgi i'm ready to lock down the feature set. give me all your crazy ideas for features you think xhip is missing IN THE SYNTHESIZER (things which the event routing system can not do.. this excludes almost all modulation things) so i can take them into serious consideration before continuing.
(no such thing as "analog sound".. it is just about quality. at this point xhip is above that level of quality in many respects, although has a way to go in others)
so now as the mp3 i posted above shows... it sounds pretty much exactly like the filter i built by a simple adjustment of the filter mod and waveshaper.. nothing really special either.
i do kind of miss some of the effects specific to a series-lossy-integrator filter (xhip has a series integrator, aka multimode) which are impossible with xhip. then again, xhip can do things impossible with a series-lossy-integrator.. so..
by the way, if you see http://xhip.cjb.net/xhip/todo.cgi i'm ready to lock down the feature set. give me all your crazy ideas for features you think xhip is missing IN THE SYNTHESIZER (things which the event routing system can not do.. this excludes almost all modulation things) so i can take them into serious consideration before continuing.
-
- KVRian
- 889 posts since 10 Jul, 2004
- a 3rd oscillator maybe? not that i ever use it but....nice to have if needed.
-
- KVRAF
- 12615 posts since 7 Dec, 2004
i'm going to just take action here (stop being lazy) and write a quick and dirty message board for xhip.cjb.net that i can moderate really strictly so we can organize better. when we discuss something like a third oscillator we can make a specific thread where all points of view are displayed clearly and everything is kept on-topic.
i was thinking maybe it could be just one main text with posts added. no need for accounts or anything. i'm working this out now, i know you guys probably wont have input but it helps to discuss this even when there isnt much discussion.. but if you do have ideas make sure to let me know!
the latest posts (30?) could be displayed on the main page while older posts can be archived. each post can be an individual text file stored on the server. i can reuse the existing directory contents display code (/upload uses it) to display the files below the latest entries.
the latest 30 files would just be included in text directly into the page. i can then go through and move/delete/whatever to organize the discussions. this is kind of semi-wiki i guess. i'm not sure what would work best.
for now i think this is a plan, any thoughts before i do this?
i was thinking maybe it could be just one main text with posts added. no need for accounts or anything. i'm working this out now, i know you guys probably wont have input but it helps to discuss this even when there isnt much discussion.. but if you do have ideas make sure to let me know!
the latest posts (30?) could be displayed on the main page while older posts can be archived. each post can be an individual text file stored on the server. i can reuse the existing directory contents display code (/upload uses it) to display the files below the latest entries.
the latest 30 files would just be included in text directly into the page. i can then go through and move/delete/whatever to organize the discussions. this is kind of semi-wiki i guess. i'm not sure what would work best.
for now i think this is a plan, any thoughts before i do this?
-
- KVRAF
- 12615 posts since 7 Dec, 2004
er, either that or i could ask ben to give me a xhip kvr board, what do people think about that? that might be better in the short term and i can move over to my own board in the future when i need more complex features.
-
- KVRian
- 889 posts since 10 Jul, 2004
yeh a kvr board would be great.
-
- KVRist
- 275 posts since 25 Sep, 2004
So..is Xhip ever going to have a stable v1.0 release with a better gui? Aciddose, I admire and respect your work, but i got bored because everyday a new feature arises and puts the project far from its "finish line". I feel disappointed when i see all these posts and i believe they will continue if you don't put a stop. I mean that a program must have some specific determined features for its first version. Then, after the completed and released v1.0, the programmers can add features or change the basics. Everything needs a measure..
It's more fun to compute..
-
- KVRAF
- 3499 posts since 9 Oct, 2004 from Poland
I'm moderately against.aciddose wrote:for now i think this is a plan, any thoughts before i do this?
1st: IMHO There is nothing wrong in using KVR Instruments forum and creating individual threads for all Xhip related topics.
2nd: I'm already a member at too many forums now and i don't wont to be forced to check another one for new posts (xhip is my favorite synth but i'm just too lazy).
[====[\\\\\\\\]>------,
Ay caramba !
Ay caramba !
-
- KVRAF
- 12615 posts since 7 Dec, 2004
algo; actually i'm obsessed with measures and very technical (verbose covering discrete topics), i try to spend as much time as i can with the actual code rather than let my obsessions control everything. if i were to spend all my time creating versioning systems and measures of progress, mapping out the whole universe and my path through it before me i'd have no time left for anything else.
http://xhip.cjb.net/xhip/todo.cgi
this is your measure that i've allowed myself to create. beta 7 is the final beta before the last implementation takes place and the final beta testing phase begins (#1).
i've always said if you're comfortable with the current features call this version 1.0 and stop visiting this thread. you might feel like we're getting further from the 1.0 version, but this is the wrong way of thinking. the todo list in my head has always been shrinking no matter what i have written for everyone else's benefit in the list on my site (#2).
a perfect example of this is the "finalize patch format" block i'm working on now. this was one meta-goal for months, and now that we're at it it's been expanded into all the sub-goals i'm aware of at this point. back when i first wanted to finalize the format i didnt know i'd need to modify the internal synth data systems to handle negative ranges for parameters (#3).
#1:
during that period the materials to be included in the final distribution will be created (banks, documentation, skins, configurations, extras) and they will then be included into one last testing phase (testing of the finial distribution) before a 1.0 release.
#2:
all the features on that list (and a few i've discarded or tucked away) have been in my head since the beginning, in some form or another. we're getting closer with every change i make, even if the list seems to be longer. elements get expanded, new sub-goals appear from the existing meta-goals once it comes time to work on them.
#3:
for example, when using the pitch bender as a modulation source on the cutoff, the cutoff may be set already to zero. this means the negative range of the bender modulation input will have no effect with the current system. i need to change the system so that the cutoff can be adjusted into negative ranges that are not available on the default gui (#4). many other parameters need this feature, and in the skins which will be produced this new system will allow much greater flexibility in the possible ways parameters can be input to the synthesizer. (#5)
#4:
abstraction of the interfaces. in order to allow many different forms of input, many vastly different styles of gui and notelogic to inter-operate with the same synthesizer i need to make my interfacing layers as abstract as possible while also making them as flexible as possible. think of this as similar to the change i made to the modulation depth parameters. first i changed the range to allow negative values. i then changed the keytracking to be full range rather than stepped. recently i've allowed individual adjustment of A/B depths rather than a simple selector. even more recently (next alpha version soon) i've made even more parameters full range. these changes are part of the synthesizer and the gui. really though they've been changes to the synthesizer. what i need to do is make the synthesizer so flexible that the gui is actually limiting it's potential to a reasonable complexity. this is exactly what i'll be doing in the next version (alpha 13)
#5:
remember, the notelogic and gui are modules separate from the synthesizer. we have three basic layers here with a few interfacing layers. each of the three layers can be changed easily. i need to make these changes as flexible and stable as possible for future software. (xhip modular? xhip 2.0? others?)
http://xhip.cjb.net/xhip/todo.cgi
this is your measure that i've allowed myself to create. beta 7 is the final beta before the last implementation takes place and the final beta testing phase begins (#1).
i've always said if you're comfortable with the current features call this version 1.0 and stop visiting this thread. you might feel like we're getting further from the 1.0 version, but this is the wrong way of thinking. the todo list in my head has always been shrinking no matter what i have written for everyone else's benefit in the list on my site (#2).
a perfect example of this is the "finalize patch format" block i'm working on now. this was one meta-goal for months, and now that we're at it it's been expanded into all the sub-goals i'm aware of at this point. back when i first wanted to finalize the format i didnt know i'd need to modify the internal synth data systems to handle negative ranges for parameters (#3).
#1:
during that period the materials to be included in the final distribution will be created (banks, documentation, skins, configurations, extras) and they will then be included into one last testing phase (testing of the finial distribution) before a 1.0 release.
#2:
all the features on that list (and a few i've discarded or tucked away) have been in my head since the beginning, in some form or another. we're getting closer with every change i make, even if the list seems to be longer. elements get expanded, new sub-goals appear from the existing meta-goals once it comes time to work on them.
#3:
for example, when using the pitch bender as a modulation source on the cutoff, the cutoff may be set already to zero. this means the negative range of the bender modulation input will have no effect with the current system. i need to change the system so that the cutoff can be adjusted into negative ranges that are not available on the default gui (#4). many other parameters need this feature, and in the skins which will be produced this new system will allow much greater flexibility in the possible ways parameters can be input to the synthesizer. (#5)
#4:
abstraction of the interfaces. in order to allow many different forms of input, many vastly different styles of gui and notelogic to inter-operate with the same synthesizer i need to make my interfacing layers as abstract as possible while also making them as flexible as possible. think of this as similar to the change i made to the modulation depth parameters. first i changed the range to allow negative values. i then changed the keytracking to be full range rather than stepped. recently i've allowed individual adjustment of A/B depths rather than a simple selector. even more recently (next alpha version soon) i've made even more parameters full range. these changes are part of the synthesizer and the gui. really though they've been changes to the synthesizer. what i need to do is make the synthesizer so flexible that the gui is actually limiting it's potential to a reasonable complexity. this is exactly what i'll be doing in the next version (alpha 13)
#5:
remember, the notelogic and gui are modules separate from the synthesizer. we have three basic layers here with a few interfacing layers. each of the three layers can be changed easily. i need to make these changes as flexible and stable as possible for future software. (xhip modular? xhip 2.0? others?)