XHip--Please finish your synth!!
-
- KVRAF
- 6242 posts since 26 Sep, 2003 from right here, as you can see ...
gsoto ... the link on aciddose's site to your gui is a 404 ...
even though i don't really believe anymore that xhip is reaching a final state ever (ad will beat me for this, i know
), i'm still curious how the gui could be looking like ...
i never loose hope, do i?
:)
even though i don't really believe anymore that xhip is reaching a final state ever (ad will beat me for this, i know
i never loose hope, do i?
regards,
brok landers
BIGTONEsounddesign
gear is as good as the innovation behind it-the man
brok landers
BIGTONEsounddesign
gear is as good as the innovation behind it-the man
- KVRAF
- 12615 posts since 7 Dec, 2004
he made some more changes to the gui and we're in a deadlock here on the gui which is due to the stuff on the start of the to-do needing to be finished. in order to implement the gui i need the numbers (pixel offsets and colors and so on) from him, but in order to give those to me he needs to have a locked down feature set. i'm pretty sure the images from a few months ago are still what the gsoto gui will look like in general, but a few things might be moved around a bit, names might be changed and other small adjustments might be made.
http://xhip.cjb.net/temp/public/gsoto_gui/
i need to fix the directory listing program on the server some day, i was messing with it and only got half-way through a while ago. here i've made a kludge/hack to get it working in this dir.
actually though, gsoto if you want to rearrange stuff and give me those numbers now i think the feature set is pretty much locked down. the additional features i still want to add will be extremely minor and wont really change the gui. i feel comfortable now saying the feature set is actually locked down for all practical purposes.
since i'm not making much progress on other things it might be a good idea to switch gears and go back to the gui, then finish the patch format stuff afterward. they actually dont really need to be any particular order anymore. as long as t he gui doesnt require multi-layer drop boxes or context menus, i've got all the controls coded and working fine already.
here is the previous quick attempt i made at implementing some of the gui. the code will be a lot neater for the finished version but this should give you an idea of the numbers i need:
http://xhip.cjb.net/temp/public/xhipeditor_autobmps.cpp - this code draws the arrow image and.. i dont remember what the other function does. anyway we'll have a file like this to load in bitmaps or generate them as needed by the gui. all this image handling code is finished and works great.
http://xhip.cjb.net/temp/public/xhipeditor_gsoto.cpp - this is the core section of the gui. the logic for how controls are handled exists here, the main window is created and functions to populate it are called. we treat all objects as containers.. so each section of the gui is actually a smaller window inside the others. each control is again a smaller window. it is best to try to balance the number of children in each parent so that the tree is balanced and it takes equal time to reach any destination leaf.
http://xhip.cjb.net/temp/public/xhipedi ... o_osca.cpp - this contains my attempt at implementing the osc.a section. the code is way too large here, i think i'll be able to optimize everything quite a bit more when i'm finishing it.
http://xhip.cjb.net/temp/public/gsoto_gui/
i need to fix the directory listing program on the server some day, i was messing with it and only got half-way through a while ago. here i've made a kludge/hack to get it working in this dir.
actually though, gsoto if you want to rearrange stuff and give me those numbers now i think the feature set is pretty much locked down. the additional features i still want to add will be extremely minor and wont really change the gui. i feel comfortable now saying the feature set is actually locked down for all practical purposes.
since i'm not making much progress on other things it might be a good idea to switch gears and go back to the gui, then finish the patch format stuff afterward. they actually dont really need to be any particular order anymore. as long as t he gui doesnt require multi-layer drop boxes or context menus, i've got all the controls coded and working fine already.
here is the previous quick attempt i made at implementing some of the gui. the code will be a lot neater for the finished version but this should give you an idea of the numbers i need:
http://xhip.cjb.net/temp/public/xhipeditor_autobmps.cpp - this code draws the arrow image and.. i dont remember what the other function does. anyway we'll have a file like this to load in bitmaps or generate them as needed by the gui. all this image handling code is finished and works great.
http://xhip.cjb.net/temp/public/xhipeditor_gsoto.cpp - this is the core section of the gui. the logic for how controls are handled exists here, the main window is created and functions to populate it are called. we treat all objects as containers.. so each section of the gui is actually a smaller window inside the others. each control is again a smaller window. it is best to try to balance the number of children in each parent so that the tree is balanced and it takes equal time to reach any destination leaf.
http://xhip.cjb.net/temp/public/xhipedi ... o_osca.cpp - this contains my attempt at implementing the osc.a section. the code is way too large here, i think i'll be able to optimize everything quite a bit more when i'm finishing it.
Last edited by aciddose on Mon Aug 06, 2007 2:36 am, edited 1 time in total.
-
- KVRian
- 673 posts since 15 Nov, 2004 from Montevideo, Uruguay
Well, acid beat me to it.
The design isn't finished at all. I keep changing it as the synth itself keeps evolving. Currently it's on its third iteration. The last posted screen is from the second prototype. It has already changed, but I'm not posting new screens yet.
The design isn't finished at all. I keep changing it as the synth itself keeps evolving. Currently it's on its third iteration. The last posted screen is from the second prototype. It has already changed, but I'm not posting new screens yet.
-
- KVRAF
- 6242 posts since 26 Sep, 2003 from right here, as you can see ...
hey aciddose and gsoto:
the dark layout looks just damned fine!
i'd really suggest (if i may) to focus on finishing ...
i know from my own experience that there will never be an end, if one doesn't force quitting at some time ...
don't you guys think it's that time now?
i don't want to sound pushy, but, like i said ... you have to define an endpoint, otherwise you'll never finish it, as there's _always_ something that could be done in a better way, or something that's missing ...
really ... finish xhip, the synth deserves it more than any other synth ... it's already outstanding ... ALL IMO ...
however, thank you guys for doing such a great sounding synth!
the dark layout looks just damned fine!
i'd really suggest (if i may) to focus on finishing ...
i know from my own experience that there will never be an end, if one doesn't force quitting at some time ...
don't you guys think it's that time now?
i don't want to sound pushy, but, like i said ... you have to define an endpoint, otherwise you'll never finish it, as there's _always_ something that could be done in a better way, or something that's missing ...
really ... finish xhip, the synth deserves it more than any other synth ... it's already outstanding ... ALL IMO ...
however, thank you guys for doing such a great sounding synth!
regards,
brok landers
BIGTONEsounddesign
gear is as good as the innovation behind it-the man
brok landers
BIGTONEsounddesign
gear is as good as the innovation behind it-the man
- KVRAF
- 12615 posts since 7 Dec, 2004
http://xhip.cjb.net/xhip/todo/
i've known exactly what needs to be done in order to finish xhip for years.
the issue is actually doing it, and that takes years.
i've known exactly what needs to be done in order to finish xhip for years.
the issue is actually doing it, and that takes years.
-
- KVRAF
- 3528 posts since 18 Apr, 2002 from British Columbia, Canada
gsoto: the screenshots of both the light version and dark version look perfect to me. personally i think you should keep it that way! 
-
- KVRian
- 673 posts since 15 Nov, 2004 from Montevideo, Uruguay
Well, I have nothing to do with the sound really.brok landers wrote:however, thank you guys for doing such a great sounding synth!
The dark version is a palette tweak by onur_77 (I have some versions by bengeorge and Kingston here too). However, I didn't put much attention to the color scheme yet, since that will be easy to tweak.
The prototype 2 is outdated, it doesn't handle the current state of the synth so I need to change it anyway.
Here is what I'm going to do; I'll try to come up with a complete design, addressing the current feature set, in not more than three weeks. Then I'll post it so we can discuss it a little and then I'll make a final version and write all the numbers and stuff to make it ready to be implemented (this will be pretty easy, acid, I can write the stuff in some predefined 'language' if you want to automate the process too).
Ok?
I have some ideas for the event routing thing but since I don't know how exactly it's going to work I won't include it in the design. I'll just destinate a whole page in the GUI for it.
I already added some new widget types but they are pretty simple (up/down buttons, 3-state buttons, etc.); I guess that won't offer any implementation problems.
- KVRAF
- 12615 posts since 7 Dec, 2004
there should be very little that will be difficult to implement. the only things which cause problems are widgets which require floating windows or complicated interactions with other controls. like i explained about the controls in a window, if there is enable/disable it is complicated to make all widgets turn grey ..
although i've solved that problem now by using my registry-database implementation. each control now is given a pointer to a database-page containing the values that should be used. when i want a widget's color to change i can just access the database page and change the value, then set it to "dirty" and the next update cycle all the widgets holding a pointer to that page will update themselves.
so, there are only a few things which will now be difficult to implement. if you just avoid using complicated interactions and weird logical associations it should be fine. try to use a flat design, top-down, controls dont talk to parents, parents 'talk to' (adjust parameters of) children controls only, peers shouldnt be allowed to communicate.
i'll try to make that very simple though, when i say "flat" i mean if you imagine a drawing of a map showing the communications between different controls, in order to get to a control you should only have to follow a straight line. you shouldnt ever have to go into a loop, or "should i go left or right?", it should always be a->z,
never: a->b or h (can go back to q) up to y or left turn toward k -> z -> h (repeat again).
although i've solved that problem now by using my registry-database implementation. each control now is given a pointer to a database-page containing the values that should be used. when i want a widget's color to change i can just access the database page and change the value, then set it to "dirty" and the next update cycle all the widgets holding a pointer to that page will update themselves.
so, there are only a few things which will now be difficult to implement. if you just avoid using complicated interactions and weird logical associations it should be fine. try to use a flat design, top-down, controls dont talk to parents, parents 'talk to' (adjust parameters of) children controls only, peers shouldnt be allowed to communicate.
i'll try to make that very simple though, when i say "flat" i mean if you imagine a drawing of a map showing the communications between different controls, in order to get to a control you should only have to follow a straight line. you shouldnt ever have to go into a loop, or "should i go left or right?", it should always be a->z,
never: a->b or h (can go back to q) up to y or left turn toward k -> z -> h (repeat again).
Last edited by aciddose on Mon Aug 06, 2007 6:22 am, edited 2 times in total.
- KVRAF
- 2841 posts since 23 Feb, 2004 from Planet Earth...for now
Actually the issue is that list of yours! It's a moving target - you're always adding things in as fast as you take them off.aciddose wrote:http://xhip.cjb.net/xhip/todo/
i've known exactly what needs to be done in order to finish xhip for years.
the issue is actually doing it, and that takes years.
- KVRAF
- 12615 posts since 7 Dec, 2004
not lately though. the last few things i've expanded a couple times (when i add things i just write the short version, no details) but the list has been shrinking fast and it is almost empty at this point.
the main problem is that damn analog synth i've been designing
http://xhip.cjb.net/hardware/pulse_cr8000.mp3
i sat down a few times to work on xhip, stretching my arms, cracking my knuckles "ok, lets finish this!".. then i say "oh shit, i should be working on that ______ module"
i've got a few people who really want prototypes so once i get that worked out it wont be so much anymore, but if i go into business making them i think for the first hundred or so units i'll be working more than 40 hours a week on that.
the main problem is that damn analog synth i've been designing
http://xhip.cjb.net/hardware/pulse_cr8000.mp3
i sat down a few times to work on xhip, stretching my arms, cracking my knuckles "ok, lets finish this!".. then i say "oh shit, i should be working on that ______ module"
-
- KVRAF
- 3499 posts since 9 Oct, 2004 from Poland
Girls !
Stop whining about the gui ! LOL
Xhip has a nice fully functional gui now and it doesn't need any shiny knobs and fake wood just like Guitarsuite doesn't need any better gui - because it SOUNDS so good and that's what's REALLY IMPORTANT.
Stop whining about the gui ! LOL
Xhip has a nice fully functional gui now and it doesn't need any shiny knobs and fake wood just like Guitarsuite doesn't need any better gui - because it SOUNDS so good and that's what's REALLY IMPORTANT.
[====[\\\\\\\\]>------,
Ay caramba !
Ay caramba !
- KVRAF
- 2910 posts since 26 Jul, 2005 from dun unda
Personally I don't mind the current GUI, although it makes some things harder to find this way, it's still usable.
But it would be neat to have a 'user-selectable skinning feature' like Oatmeal.
But it would be neat to have a 'user-selectable skinning feature' like Oatmeal.
-
- KVRian
- 795 posts since 2 Apr, 2006
That dark gui looks the works.
(PS: if I try to open the gui on the last linux version, it craches ext2, instant close
I can only design sounds by automating on the parameters to the values I want from automation lanes, not really handy)
(PS: if I try to open the gui on the last linux version, it craches ext2, instant close
