here's a small update. slight changes here and there such as making it fit at 1280x1024 and a few tweaks. see notes in the source file's head for details. build for 6.2 included.
r-a-c.de
The collected works of foetz - Page 23
vishnu wrote: Wow, thanks! Did you run a code prettifier on it? Because if you did it must be one that likes long lines because the line in reset() that rezeros all the labels now goes out to column 4,648...
haha yes, partially. the long lines shouldn't be a problem tho for the compiler or the user thanks to line wrapping all decent editors have
Also you prefer to include stdlib.h instead of cstdio? Although I suppose on IRIX 6.2 that makes sense.
exactly, i had to on 6.2. that was before c++98 so no namespaces either and such. but that's of course not meant to be better for all compilers and system having the "official" c++ headers.
Should probably ifdef that and the STL3 includes with _sgi and if it's not defined include the regular ones.
well as i wrote in my comment that's just how i called that dir on my box. there was no stl at all for 6.2. it had to be downloaded by the user and then you could put it anywhere and call it anything. having a more recent os as target that's of course not practical but was the only way getting it done on 6.2.
And the <iostream> stuff was just in there for debugging from the console, if we're not using it the setValues() count variable is unecessary. And since X obviously won't let us watch all the numbers fly up individually I should remove all the calls to XtSetValues inside the simulation for() loop, that would take care of the ridiculous 5 minute lag...
hehe yeah sounds good
EDIT: Still need to include <iostream> for the call to time() in the random number seed...
oh seems i missed that then
r-a-c.de
vishnu wrote: Done! The whole thing now runs in a fraction of a second.
great
I ifdefed the 6.2 parts so it should compile on both IRIX 6.2 and Linux
see above
you changed my use of XtOpenApplication() to XtVaAppInitialize() which is deprectated in X11 7-point whatever, but I left it.
oh sorry, i didn't know. if there're any benefits please do change it again by all means. sorry for the extra hassle
r-a-c.de
and here comes v1.3
array increment removed, fixed one new bug and added a timer to simulate().
working like a charm now without the instant label updates. the whole thing takes less than 2 seconds on an r4400 indigo2
updated source and the resulting build for 6.2 included.
array increment removed, fixed one new bug and added a timer to simulate().
working like a charm now without the instant label updates. the whole thing takes less than 2 seconds on an r4400 indigo2
updated source and the resulting build for 6.2 included.
r-a-c.de
nongrato wrote: I thought I tried it with gcc. I'll check later.
yes, the error output you posted is gcc
r-a-c.de
i use my sgis for whatever they can do. for the rest i use osx.
as for the challenges (pun partially intended
), i don't think that changed much compared to back when they were still sold. they were never intended to be consumer allround machines so there's always been something missing here and there
as for the challenges (pun partially intended
), i don't think that changed much compared to back when they were still sold. they were never intended to be consumer allround machines so there's always been something missing here and there
r-a-c.de
if that's a discreet setup then the desktop is limited. they turned off as much as possible since the machines were delivered as turnkey systems
r-a-c.de
looks ace
especially the water drops are great. i almost don't dare to say that there's one tiny detail that looks not as perfect as the rest which is the foot. unlike the very top that looks a bit to smooth and blue for my taste for being the usual can metal
especially the water drops are great. i almost don't dare to say that there's one tiny detail that looks not as perfect as the rest which is the foot. unlike the very top that looks a bit to smooth and blue for my taste for being the usual can metal
r-a-c.de
commodorejohn wrote:vegac wrote: Pity Dillo is so limited...it's certainly got speed on its side.
Unfortunately, the fact that Dillo is so limited is why it's got speed on its side. We put such absurd demands on web browsers now, with elaborate layouts and absolute mountains of Javascript even on a lot of pages that would be perfectly fine without it
couldn't agree more. here's a random example that makes one think wtf?
:
r-a-c.de
Dans34 wrote: attempt to climb on it
r-a-c.de
maybe someone had that issue in the past and found a solution. i never did, didn't care much either but it never stopped bugging me so here it is:
some of the old scanline renderers such as wavefront's image or prisms' crystal don't show the colors right. i didn't have to match colors often so it wasn't an issue mostly but just recently i found out by accident that both show the same difference from the correct color so i figured that might be something general.
i attached 2 images, one from wavefront (scanline) and one from poweranimator (raycaster). now the thing is for both materials i used the same rgb/hsv values and matched whatever else the programs provide. you can clearly see that the color is different. it's not a matter of lighting or something else but it's simply not the same color.
i tried different machines, different irix versions and different version of the programs but always the same result. and it's not that the renderer would run in some sort of limited color mode. if i adjust the color values a bit i get a fine match. the crucial incident was when i noticed that the old prisms renderer had the same color difference so the big question is why?
what do these scanline renderers have in common that's causing the same color difference? must be something general since it's affecting both and they're not related in any way.
and fwiw, even when i use mantra in prisms and set that to scanline the colors are correct so it has to be something exclusive to the old style renderers.
some of the old scanline renderers such as wavefront's image or prisms' crystal don't show the colors right. i didn't have to match colors often so it wasn't an issue mostly but just recently i found out by accident that both show the same difference from the correct color so i figured that might be something general.
i attached 2 images, one from wavefront (scanline) and one from poweranimator (raycaster). now the thing is for both materials i used the same rgb/hsv values and matched whatever else the programs provide. you can clearly see that the color is different. it's not a matter of lighting or something else but it's simply not the same color.
i tried different machines, different irix versions and different version of the programs but always the same result. and it's not that the renderer would run in some sort of limited color mode. if i adjust the color values a bit i get a fine match. the crucial incident was when i noticed that the old prisms renderer had the same color difference so the big question is why?
what do these scanline renderers have in common that's causing the same color difference? must be something general since it's affecting both and they're not related in any way.
and fwiw, even when i use mantra in prisms and set that to scanline the colors are correct so it has to be something exclusive to the old style renderers.
r-a-c.de
nice. now we just need the old theme back because the default here doesn't work well with neither dillo nor most mozilla versions for irix
r-a-c.de
yay that was quick, much thanks! as if nothing had changed
imagine the irony letting this of all places become unusable for older machines/browsers
out of curiosity, did the previous version cause trouble? or some security issues?
imagine the irony letting this of all places become unusable for older machines/browsers
out of curiosity, did the previous version cause trouble? or some security issues?
r-a-c.de
found a bug already
on the notification options page, turning off notifications for new pms doesn't work even if email is selected, too.
on the notification options page, turning off notifications for new pms doesn't work even if email is selected, too.
r-a-c.de
diegel wrote: I think it would be a good idea to make the subsilver2 theme the default.
indeed. at least for existing members and after signing up
r-a-c.de
yeah gamma seems to be the best bet. even more tricky with wavefront because although setting a different gamma for the renderer wavefront's own image viewer ignores it so although the rendering is right you only see it if you're not using wavefront's viewer or set the gamma for that, too
and not enough it also has that "wrong" gamma applied to the material editor so getting it right for a different target gamma seems nasty. anyhow those who know wavefront know that it's an env var paradise so there might be one for the gamma, too
and not enough it also has that "wrong" gamma applied to the material editor so getting it right for a different target gamma seems nasty. anyhow those who know wavefront know that it's an env var paradise so there might be one for the gamma, too
r-a-c.de
josehill wrote: That's exactly how I operate. My main Neko bookmark is the new posts one.
GL1zdA wrote: Personally, I always start with "New posts" so I've simply added search.php?search_id=newposts - to my bookmarks and Firefox start page, this way FF always suggest the correct link whenever I start typing "neko".
Same can be done for "Active topics": search.php?search_id=active_topics
and the same goes for me but the point is what Trippynet said
r-a-c.de
josehill wrote: I should add that the way I read Nekochan posts most often is via an RSS feed reader . I find the RSS feed to be the fastest way, by far, to keep up with what is being posted in the forums.
rss is nice but only when using it through a local program. otherwise in case of web based services you give them all your feeds, they see what you read when and for how long and from where and so on ...
r-a-c.de
mozilla used to have an rss reader included. it could be compiled with the main program as an option. i remember including it with some of my builds back then. it looked quite similar to the email client. not sure if it's still around tho
r-a-c.de
it seems seamonkey still has it as part of its email client
r-a-c.de
my pm counter shows -1
anyone else having that?
anyone else having that?
r-a-c.de
ClassicHasClass wrote:foetz wrote: my pm counter shows -1
anyone else having that?
I ate your PM. It was delicious.
r-a-c.de
Knuckx wrote: They did not want to unpack from their zip file(s) though; I ended up using Universal Extractor, and even that complained...
same here. i tried different programs and names but no luck with whatever i did
r-a-c.de
Alver wrote:Code: Select all
$ mv 5014-1449_1.z01 5014-1449.z01
$ mv 5014-1449_2.z02 5014-1449.z02
$ zip -s 0 5014-1449.zip --out 5014-1449-unsplit.zip
$ unzip 5014-1449-unsplit.zip
what a mess, could they have done it any more twisted?
anyway thanks of course for the links and the "how to extract" guide
EDIT: no idea what version of zip you used but neither debian nor osx could handle it.
EDIT2: ended up concatting them manually and throwing that at Universal Extractor. no other program i had could handle it. a very weird way of packing something
r-a-c.de
well i used all sorts of packers that can handle multi file zips and none could unpack it. anyway thanks to Knuckx for the Universal Extractor tip
r-a-c.de
vishnu wrote: The Dillo mailing list has been humming like a hive the last few weeks, not to give out any spoilers but the most active topic has been "what's currently broken in dillo"...
not a bad thing to hunt down probs. another thing to get sorted is the ssl warnings. they either need to add a "don't ask again" checkbox or keep the output to the terminal. i removed a bunch of them for my 6.2 build and i reckon for this one diegel did so, too.
oh and they need to correct the changelog. it does not work with mipspro but only gcc.
r-a-c.de
okay one other issue, editing sent pms gives me "The submitted form was invalid. Try submitting again.". happens when hitting preview as well as sending.
working fine for new pms and replies
working fine for new pms and replies
r-a-c.de
for a fresh install you need:
then if you wanna take the easy route:
if you don't wanna swap cds all the time copy all the inst packages into one dir and point inst or swmgr to that
Code: Select all
IRIX 6.5 Foundation 1 CD (812-0759-002)
IRIX 6.5 Foundation 2 CD (812-0760-002)
IRIX 6.5 Development Libraries CD (812-0766-002)
IRIX 6.5 Development Foundation CD (812-0757-002)
IRIX 6.5.x Overlays (1 - 3) CD
Compiler Execution Environment 7.4 CD (812-0924-002)
MIPSpro C compiler 7.4 CD (812-0707-004)
MIPSpro C++ compiler 7.4 CD (812-0400-010)
MIPSpro Fortran 77 compiler 7.4 CD (812-0403-010)
MIPSpro Fortran 90 compiler 7.4 CD (812-0402-010)
MIPSpro compilers and tools 7.4.4 CD (812-0980-007)
then if you wanna take the easy route:
Code: Select all
Inst> keep *
Inst> install standard
Inst> go
if you don't wanna swap cds all the time copy all the inst packages into one dir and point inst or swmgr to that
r-a-c.de
oh what a nice time travel
hmm actually, since you were the one ripping it, could you upload the original somewhere? not sure if youtube applies some additional conversion
hmm actually, since you were the one ripping it, could you upload the original somewhere? not sure if youtube applies some additional conversion
r-a-c.de
once i had a disk with similar issues. i ended up replacing it.
r-a-c.de
pentium wrote: The video was uploaded as an uncompressed but deinterlaced AVI (my ISP must of loved that because it's nearly seven gigs).
that doesn't sound like uncompressed at all. it's 32min at 480x360@30fps. uncompressed one second is 16mb so one minute is almost 1gb already.
anyway pity you let youtube do the encoding. something like that should be preserved in better quality even if the source isn't the best either.
r-a-c.de
fair enough and thanks again for posting
r-a-c.de
6.2
this one doesn't need libssh2 which is more handy and fixes one issue i had with the makesfiles i.e. they override the -O options provided by the system's CFLAGS in case you run configure with --enable-optimize
- curl_7370.tardist
this one doesn't need libssh2 which is more handy and fixes one issue i had with the makesfiles i.e. they override the -O options provided by the system's CFLAGS in case you run configure with --enable-optimize
r-a-c.de
zfs needs more work. have a look at:
http://waddles.org/content/replicating-zfs-root-disks
r-a-c.de
hamei wrote: Mary, Joseph and the Baby Jesus ... when I installed Sol 10 I made the root disk ufs on purpose but the new installer doesn't give you that choice.
yeah that's what's called "progress"
What do you think about booting from a CD and using dd ?
i've never been a dd friend. first and foremost because it's dead slow. also zfs might store something like a unique id per disk or something that'd make the clone useless.
r-a-c.de
TeamBlackFox wrote: I know C but not enough to dabble in a 20+ year old makefile and source code and make changes.
the age doesn't matter. many things were actually easier. if you can handle c and makefiles today you'll have no problems dealing with older stuff
r-a-c.de
uunix wrote: You gotta love Indy..
Mines doing work tonight also.
oh you're running cde. that's quite a rare sight
r-a-c.de
josehill wrote:foetz wrote: oh you're running cde. that's quite a rare sight
I'd boot mine up in CDE every now and then, just from curiosity. As far as CDE experiences go, the SGI version was quite nice and snappy. I can't quite put my finger on any objective reason why this was the case, but CDE somehow felt smoother on an SGI machine than on Sun or AIX machines of similar vintage and power. It's easy to setup CDE and then switch back to the Irix Interactive Desktop when finished (just a matter of installing the CDE eoe and then setting/unsetting some chkconfig flags), so if you are at all curious, I suggest giving it a try.
oh i did run cde on irix myself a couple of times. it's just not something you see often
r-a-c.de
josehill wrote:uunix wrote: The desktop you see is from a Power6 AIX CDE, I was telneted into the Indy.
...which makes the last few remarks by foetz and me even funnier. Well, at least to me.
for sure, same here. so hey, seems we have an advantage then
r-a-c.de
kjaer wrote: Inconvenient software wasting time is a long way from boilers exploding and killing people. Things improve when people are of a mind to get together and improve them.
things improve when one or more people have determined what improvement is. and that's the problem, the idea of what improvement is is very subjective
r-a-c.de