The collected works of foetz - Page 22

yes, that was digital domain's in-house compositor but i've never seen it on irix either

• gimp_125.tar.bz2

never hurts to have a classic. the last gtk1 version including the data extras that were available back then. it's a separate dir for the whole thing that goes to /usr/local so that you can run gimp2 and this one one the same machine without problems.

hamei wrote:

foetz wrote: ...

foetz, can you get Gimp 54 to File ... Open ... anything ? I've tried jpg's and tiffs and pngs with no success so far No nastiness, just silently fails to open the file. I can make new stuff but not import anything.

seems you don't have a wildcard for the current dir in your $PATH. either add the gimp dir or the general .
indeed that's just a dot. example:

but either way as i added to my other post it doesn't do what you want. it just shows the crop dimensions but no entering own values let alone a fixed retangle. you might wanna have a look at this: viewtopic.php?p=7374452#p7374452

and the sgi related stuff in the makefile seems to be from a campus network they used. not sure what they used there tho because the code and especially the build system is utterly crappy

haha looks like a real workshop at your place. the power series logo in the back should put you in the right mood

jan-jaap wrote: That's not (just) a logo

yeah i didn't mean it like that. i did notice it's the actual thing

great thanks. gonna be fun having the original elements of something i used so much. still do actually for email

well welcome back then! always good to have the "old" guys around ... or coming back

i never felt the need for an ssd. i simply don't need that speed neither want that kind of "reliability"

hamei wrote:

nongrato wrote: HIIP from the Hotmix#18 CD ...

Eighteen, you say ?

it should be on the cd from avid or whatever other source you have since it's needed to run the thing

vegac wrote: Had some code projects go from 45 minutes down to 5 minutes for clean builds.

ramdisk

5.3

• climm_071.bz2

6.2

• climm_071.bz2

the classic icq client without jabber and other fancy stuff

• climm_071.tardist
• climm_071_ssl.tardist

the classic icq client without fancy stuff like jabber or otr

serious backups shouldn't be running with the target machine. hook the medium up, run the backup, disconnect, put away. at least as far away so that any sort of accident of the box can't affect the backup.

sgtprobe wrote: Haven't seen this one before. SI3D V1.0 demonstraded at Siggraph 2006 on an Octane! Quite hillarious

https://www.youtube.com/watch?v=nzj3VITdN5A

we had it here before but i couldn't find the thread right now. but as i said back then that's a video from douches for douches
including the one who put that title up because si3d v1 wasn't called si3d and wouldn't run on an octane either so that dude has no clue what he's talking about.
same seems to go for the peeps in that video. treating the founding fathers of 3d that way is equally dumb and ignorant. if i would shoot a parody video about windows and x86 (and xsi) it'd easily fill several seasons of "big brother" and that'd be actual, serious issues contrary to what they picked up in that video.
especially given that xsi is by far the most crappy 3d package so far. only runs properly on winblows and even there stability and features are humble compared to others; modeling in particular. i remember at a seminar the instructor from avid couldn't get the network render to work even once all day and one of the default scenes crashed badly. the linux and irix versions were just junk
they had it coming, xsi lost a lot of ground and the only serious role it plays anymore is for games. and even there others are more popular. compared to the days of the original si3d its market shares are just a fraction today with no silver lining on the horizon in the near future. even today i'd prefer an sgi with si3d over any sort of xsi anytime.
i guess going down that road in the video was their way of dealing with the sad reality because even then they were headed south already. bad style.

sgtprobe wrote: By the time the demonstration of SI3D in the youtube clip was made (2006), XSI was so far ahead in every possible way that stating anything else is just fanboyism, sentimentalism or pure ignorance.

and that sounds remarkably the same

of course my post was to be read with a grain of salt. i do realize that every supermarket peecee beats the crap out of every risc box from back then but that wasn't the point.
in recent years i got quite fed up with kids (or older but ignorant peeps) thinking that the new and fancy stuff is much superior compared to anything else just because it's new and fancy which in return makes them cool, too. in fact some quite strict rules have evolved regarding what's to be considered superior and widely accepted. that's also not to be questioned and if you do so, even if you're right, you're outlawed instantly. thinking outside the box is not welcome and that's a bad sign.
si3d with an sgi was a mature set for pros and something like that doesn't exist these days. that's sad and even more sad is having ignorant people making fun of that while sitting in front of their off-the-shelf junk thinking they're the greatest because they have a github account and praise python that's a bit like a driver of a new toyota patronizing the owner of an older bentley just because he's got more horsepower ... you get the idea.
is that sentimental or emotional? of course it is because as a human being i can't be 100% objective nor can anyone else. but that doesn't mean it's not true and that's the very point.

as for xsi in particular, the flaws i described were no secret at all and not related to the guy behind the keyboard either. gocram said it already, especially in the beginning features and reliability were so poor that they not only released si3d v4.0 but also gave you one for free with every xsi purchase to compensate what xsi messed up or couldn't do at all. i've been in contact with avid at the time and they didn't hide it so that's officially confirmed. don't get me wrong xsi did have some nice stuff but it never was a mature package like its precursor.

sgtprobe wrote: Ok, fair enough! I do understand what you are aiming at here, and I have to agree.

I hope I didn't came off too harsh, because it wasn't my intent. Everything is so much easier to discuss in person over a few good beers, posts on the forum can look a bit more aggressive than they are (reading what I wrote). So, I'm sorry if you think I went over board!

no worries at all, i knew that post would polarize a bit which was intended

But I don't understand why it isn't a mature package? Have you even seen XSI from version 7 and forward?

i have, 2012 was the last one tho i reckon.
by "not mature" i meant it wasn't/isn't complete. especially the modeling never reached a complete state and they were not interested in that either. at quite an early stage the roadmap put most emphasis on animation and as it turned out they kept it that way till the end.
then there were other things like ice that made me chuckle because it was such an obvious houdini copy. not only the concept but also the name since prisms' compositor was called the same was quite looking like desparate measures at the time already ...
... and so on. although i've never been an avid softie (<-- double pun ) i did respect si3d for what it offered and that just never happened to me in case of xsi because of stuff as described above. in addition the fact that it was pretty much a pure windows app didn't help

ultrix does come with x. if you wanna replace the native x you have to consider a lot of things and might still break some. obviously the first step would be compiling it on ultrix and then carefully replacing all parts of the original one and checking all related scripts you can find for incompatibilities. then you could hope for no serious issues and take it for a first ride

sgtprobe wrote: Maya sure looked tempting ... it really felt fresh and new and all that jazz.

yeah but that's about it, maya wasn't great at first either. poweranimator was beating the crap out of it modeling wise for many versions to come and maya's default renderer sucks (present tense because it never got better). the gui was fat and slow and the mentioned renderer only had smp support for some parts. selection of overlapping lines or points was pure luck compared to the oh so handy popup late poweranimator versions had for such cases ... and so on.
you see i'm not interested in bashing xsi just for the heck of it, i could just shake my head everytime i see new stuff coming out that's miles behind what's there already and xsi was just one of those.

But I guess they (Avid) saw Wintel being the next "big thing", which, sadly, it was, not that it helped XSI in the end though.

in fairness that wasn't all up to avid. when xsi was started (called sumatra during the initial development phase) it was still under microsoft's roof and they did of course choose windows as their primary target. i'm actually not sure there would have been versions besides windows at all if microsoft wouldn't have sold softimage.

Today I'm working on your typical (well, maybe not, loaded with graphics hardware for GPU rendering and such) Off the shelf hardware running on Winblows (as you said). I start her up in the morning, working/rendering/modelling without any issues as a user without actually worrying what it says on the box. It's a tool, and it get's the job done and fast (although it could always be a gazillion times faster, but that goes for everything).

hehe yes sure and that's perfectly fine of course. today there's no other choice even if you wanted something different

lucky7456969 wrote: No, probably not

do you have a desktop or not?

rvic wrote: Just to be clear from the start, i thank you for your work and i have waited for many apps like these to be available for lower version of 6.5.22.
...
I want to express my gratitude for your work ... just to be clear ... i just need a clear path to enjoy them

no problem at all, let's have a look ...

After i somehow installed dillo3, it's a mix of tgcware and your libs and i ended up with segmentation fault ...

not as it should be

Well ... the problem is ... tgcware is compiled against /usr/tgcware/lib at least that's what i assume
Your 6.2 package installs in /usr/local/lib but, for example, dillo3 doesn't look in there, so i had to put ld_libraryn32_path in my cshrc ... but i already know from experience that is a bad idea.

Then ... dillo3 needs gcc3.46 from tgcware that's linked to some 6.2 patches and some libs from tgcware:
listdcxx newest version and it's strange considering that gcc3.46 is the oldest package, also libgcc newest version same argument.
In the end libiconv is duplicated (tgcware plus 6.2)

okay, first of all using environment variables like LD_LIBRARYN32_PATH is not a bad idea. in fact those variables enable great flexibility and freedom ... if used right.
then you need to be aware of the rulesoverride option for installations. have a look at this: http://techpubs.sgi.com/library/tpl/cgi ... 936-PARENT

the solution for your problem is using rulesoverride and only installing ONE of each lib. for example if you have a tgc libiconv already you don't need mine or if you have my libpng you don't need the tgc version and so on.
the major difference between tgc and my packages is that tgc is completely built with gcc while i use the original sgi compilers and dev tools. i do make exceptions tho if the targets use very new or very crappy code
you can mix programs and libraries built with gcc and the sgi compilers unless they're written in c++ and since almost all basic stuff makes no use of c++ you should be fine mixing them. so clean up your system and keep it to only one version of each lib and keep the library paths. then things should be fine

attached package contains the iv files converted to obj and sidefx format

sgtprobe wrote:

foetz wrote: attached package contains the iv files converted to obj and sidefx format

Absolutely fantastic! Thanks a lot for this!

no problem at all. was just a few clicks

sgtprobe wrote: Base done in Psubs, but I don't do any weighting of the edges , so I guess it could be done in ordinary subdivs anyway.

alternatively you could template the original stuff and put a few curves on

oh yea how nice. seems the at&t guys finally got it right

i actually prepared this for christmas but since the forum was down i'm just posting it now ...

it always bugged me that there was no proper ebay proggie for anything but windows. yes there're some java based tools but java ... oh well. and even those didn't run well on unix or even not at all. other implementations were rather humble and especially for those running an sgi server something like that would be very handy instead of having the desktop running with the sniper thing until the auction is finally over.
so as a little xmas special this year i figured i could share one of my own: RACbay, an ebay sniper.
should be quite obvious what it does and what it's for but just in case here's a short show off list

• what it can do: bidding on auctions
• what it cannot do: everything else
• it's compact: just 120kb (dso version)
• it's compatible: runs on irix4 and later
• it's fast: easily beats the java based tools even on irix4
• it's bloat-free: because it runs on irix4
• did i mention that it runs on irix4?
• it has 2 styles: motif classic and indigo magic. switchable at will (irix5 and later)
• it's discrete: not sharing your data in any way. no cloud or feedback or other crap like that
• unlimited auction count (well, as many as unsigned int can handle )
• straight forward use
• brief and clear log messages
• does not rely on nor alter your system clock
• builds for all irix versions from 4.0.5 to 6.5.30; static and dso. database and config files are interchangable
• switchable use of either the .de or .com site

now of course there's a catch
it's using the ebay api and therefore my dev id so i'm not gonna provide a public download. if you're interested just pm me and you shall have one. i will however take the liberty of rejecting requests if for some reason i don't wanna trust the guy with my dev id. e.g. if you just joined or made 4 posts in 6 years then i have to decline politely

and just in case for the record, this is in-house software without any warranty whatsoever. it's not open source and the user is not entitled to anything. i'm providing this for free and just from the good of my heart - uugh sorry, okay that got a bit too cheesy but you get the idea


EDIT: Tru64 version added

5.3

• ncurses_57.tardist
• libiconv_1111.tardist

6.2

• libiconv_112p1.tardist

the libiconv for 5.3 is a replacement of the previous one. it now comes with shared and static libs.
same for the 6.2 version but there the static libs are just a patch for the original tardist so you need both.

uunix wrote: does it matter if you logging ebay.co.uk rather ebay.com?

no, what matters is where you bid i.e. buy. if you bid through ebay.co.uk and win the auction then all further actions such as complaints, payment and whatever else have to go through ebay.co.uk. no matter where you're from.

and thanks of course for the compliments

vishnu wrote: Amazing! Is it a Viewkit app?

it works on irix4 so no, it can't be. it just has 2 different sets of x resources. run it with option -s and you have the sgi look. otherwise the motif default i set.

If you stripped out your ID stuff then wouldn't you be happy to let us have the source code? We could graft in our own user IDs, compile, and it'd be good, yes?

hehe maybe one day but for those who have their own ebay dev account i could add an option. however those users are a minority.

And to think I thought I was busy during the nekochan downtime writing a program to recreate Julian Braun's "how many time you will get each hand vs. the dealer's upcard" in 100,000 hands of casino 21. Braun was working at IBM at the time, it was the 1960s, and he wrote the code in fortran. Needless to say I haven't written a line of fortran since college so I'm using C++...

ha nice, a hobby thing or do you work for one of these poker sites?
anyway sort of a pity to replace fortran with c++ for a calculation proggy

vishnu wrote: As far as I've been able to determine nobody's got Julian Braun's original fortran code

yeah i just had a look myself and all i found was a newsgroup post from one guy who said there's no digital copy. anyhow fortran is very good when it comes to backward compatibility so assuming it was something like fortran II or IV just a couple of changes might be enough to make it work with more recent compilers. after all stuff like that is exactly what fortran is for

does it show that i'm a fortran fan?

please tell us how you set the network up on the alpha. e.g. local ip, gateway ip, dns ip ...

first assign a static ip. then make sure you have the right default route and then try again

uunix wrote: There must also be a resolve.conf

there is

diegel wrote: Many problems I had with the firefox3 build caused by broken mips pro nekoware packages

which isn't mipspro's fault.

anyhow as long as no c++ libs are involved things should work together. you could however add something like -gcc to the file name or so so that people don't have to download first check the readme for finding out whether it's gcc or not.
the important question tho is how many of the programs that don't work with mipspro do actually make sense on irix? the vast majority of the so called "modern" stuff is made on and for linux and in many cases is bloated as hell to somehow compete with the commercial stuff available for windows and osx which they don't accomplish in the end anyway
in fact a fair share of stuff doesn't even compile with anything but gcc on linux which is so poor. i had hopes that the increasing popularity of clang would change that a little but that didn't happen yet ...

anyway getting too far off topic now. any kind of contribution is welcome of course, even gcc builds

good to hear, have fun

diegel wrote: I can't see any reason to build programs written for gcc with mipspro

not just gcc but gcc/linux/x86. what might make sense for gcc/linux/x86 could be not so great for let's say gcc/osx/ppc or gcc/solaris/sparc ... you get the idea. the compler is only one part of the whole thing.

either way you nailed the problem, writing stuff for one specific compiler (and os in worst cases) is pretty much the worst case except for very special situations.

diegel wrote: The only way to change this is paticipating in open source development. Don't blame a community that don't has access to our compilers/systems

not at all, this isn't limited to mips or any specific platform. the opensource community has free access to clang, sunpro for linux and solaris and the intel compiler for non commercial use and pgi had a free offer for osx as well not so long ago ... and others i forgot about right now
there's no reason for limiting themselves to gcc and certainly nobody forced them to "abuse" the *features* of gcc. by simply running the code through a serious compiler one can discover many flaws the source might have and start delivering an actually proper product.

yay you did it hehe. gonna try it later today

hmm, what's wrong with hp-ux? last time i checked it was fine and recently hp said they wanna keep it up for another 10 years

diegel wrote: Is there any freeware library for hp-ux comparable with nekochan?

indeed and even a professional one. but see for yourself: http://hpux.connect.org.uk

ClassicHasClass wrote:

foetz wrote: hmm, what's wrong with hp-ux?

Would you like that list alphabetized or by priority?

just tried it on osx and it's working fine. irix coming up ...

ha, fantastic
that needs to be archived. here's the package just in case: