On Fri, 3 Mar 2006, Tim Freedom wrote: > Hi, I've posted about GTKwave in the past and can't really put my > finger on what was agreed upon and/or what the future roadmap for > the project is. I'm not entirely sure where GTKwave resides (in > terms of a homepage) and not at all sure who is working on it and > which version to use. The various links are either dated, point > to the wrong info and/or contradict each other. For instance, > > - GTKWave's homepage (?): > http://www.cs.manchester.ac.uk/apt/projects/tools/gtkwave/ > NOTE: No progress since May 2004; is project dead or are they > shopping for a new owner ?
http://home.nc.rr.com/gtkwave/ is the latest 1.3 version. 2.x has ceased development but 1.3 continues. There is a lot of stuff in 1.3 that 2.x does not have (scroll mouse, LXT2/VZT support, VCD export, analog support for variable height traces, etc). > - Win32 GTKWave: > > http://www.geocities.com/SiliconValley/Campus/3216/GTKWave/gtkwave-win32.html > NOTE: using version 1.3.19; last update June 2002 See the roadrunner webpage above. Latest one up should be in the 1.3.80s. I don't have a current version up as a buffer exploit hijacked IE and wrecked my windows XP install. Reinstalling XP is not a priority of mine. Them's the breaks. It should fairly easily recompile under MinGW. > - gEDA GTKWave: > http://www.geda.seul.org/tools/gtkwave/ > NOTE: Current version info (2.0.0pre3) seems dated, no mention of > current status and/or what is happening in this application. > NOTE: The 'Official GTKWave homepage' link is not really pointing > the ATP's GTKWave's page > NOTE: The 1.3.x series has progressed to 1.3.85 (webpage notes 1.3.59); > so are we to use the 2.0.x series or 1.3.x - which is newer and > why the schism, will there be future development, whom to > direct questions/comments to, etc. The "schism" is because the people working on 2.x added async support for Balsa so as far as I can gather the tool does what they need for them. Plus everyone working on it has gotten their Ph.D's and is probably gone from the uni. > GTKWave is an __extremely__ important tool for anyone doing RTL design > yet it is a bit cumbersome to use (2.0.0pre3 at least is) especially > its "Signal" search and hierarchy tree. It would be ideal to make it > a bit more user friendly and intuitive akin to Signalscan, SimVision, > Virsim or Debussy (they all seem to follow the same intuitive segmentation > of hierarchy and signals which ought to be studied). I'm sure GTK as > a library has something that can be tapped into to help out. Compile and install the latest version using gtk+-2.x. The tree search is much nicer now. (It was reworked by Tristan Gingold of GHDL.) > Is there simply no interest (or need) in further clarifying GTKWave's > current status and future plans ? Are there no needs to further develop > (or enhance) this tool ? Development continues on a daily basis with user contributions and also from what I need featurewise in order to do my job at work. As it stands, there is no roadmap or anything like that as development isn't done by committee. I'd really like to add sourcecode annotation like Debussy does but it's not a priority of mine as I've gotten along fine without it so far for debugging fairly large SoCs. gtkwave should be on sourceforge sometime soon as there's another project that sort of integrates with the viewer (OpenSKIL) and the maintainer would like all the source in one convenient spot. I actually received an email from him today about it. Regards, -Tony
