The collected works of jimmer - Page 3

TeamBlackFox wrote: I hadn't really thought of SPARC though... does anyone here have firsthand experience with a box that would be usable as a daily driver? [...] but my Ultra 5 was a huge disappointment in the performance department so hence why its not listed in my lineup anymore.

I dont have experience with recent SPARC hardware but 25 years ago I lurved my old SPARC stuff: SS2, SS20, IPX. My experience aside, it's clear that your RISC options are limited to SPARC, POWER or PA-RISC. You asked for something else than POWER, I gave you SPARC

TeamBlackFox wrote: I'm done with x86 hardware dying and costing me $300, $400+ to replace. This box lasted a scant three weeks of being on 24/7, that is just pitiful. I'm not willing to continue to play Intel and AMD's games of planned obsolescence either.

Interesting, my experience have been quite the opposite. My AMD Athlon II X4 620 cpu with stock cooler on an ASUS 760G-based motherboard in a no-name pee-cee case 24/7 has been running trouble-free for almost 5 years. I bought it because it was the first sub-100-units-of-money-quad-core CPU, and I figured that four cores at 2.6GHz would last me at least 5 years. They have and to be honest for my needs it's still fine.

Nonetheless, I've been eyeing the Dell XPS 1810/1820 or considering having a go at casemodding a fanboi 'modern' SGI Workstation. But I digress.

TeamBlackFox wrote: Yeah I avoid GPL when reasonable alternatives exist. If I had to choose between IRIX GPL or closed I'd choose closed.

If you carry on like this, your pet troll will have to go on a crash diet...

@canavan - thanks! will apply this with the next release of Pretty.

Hi All,

Am looking for a clean set of O2 skins.

As I have reasonably clean grey top plate and PSU cover, I'm also interested in hearing about partial sets consisting of only the grey base and blue shell.

UK location preferred.

J.

Hi All,

Here's a gcc 4.7.1 compile of GraphicsMagick 1.3.20 (August 2014 release, default compilation settings: -O2) versus stock dmconvert.

Benchmarking of any kind is notoriously unreliable and I doubt anybody cares about IRIX and speed in any real world setting anymore, but the speed difference here seems quite 'real'.

Should we be looking into systematically replacing certain IRIX libs and utilities with current FOSS versions?

Self-Important. Agree. Unwarranted. Disagree.

I do not know if RMS has made any relevant code contributions to a recent FOSS project and we can debate Hippy Idealist vs. Corporate Suit till the cows and the shareholders come home, but I would suggest that from a historical viewpoint few programmers have had a comperable impact on their field.

@khalidschofield - sorry for late late response. Pretty is a bit blond I'm afraid, there's no actual binary and no command-line options. It's just a script that pops-up a splash on install, copies some background images to your machine and then modifies your backgrounds index file so you can set the copied images as a background using the stock background control panel. (Desktop->Customise->Background).

Well, Pigs flew - Villa is off to Wembley

/usr/local FTW!

Have fun at Wembley - and Good Luck!

Also, from what little I understand and/or know from hardware, it seems to me that SGI MIPS CPU's are significantly different from the modern MIPS designs making it non-trivial/impossible to use modern MIPS compiler back-ends to generate code for older SGI MIPS CPUs.

People who actually know what they are talking about might want to correct my statement though.

Nostalgia! page 254 - Olivetti Quaderno - I had one of those. Such a fun little thing

ivelegacy wrote: About Irix: i'd like to see "geany", as I feel I miss an useful IDE. [...] Geany is a powerful tool, unfortunately has to deal with GTK+, so … it may hurt hard. I am used to use it under linux just because I do not have to care about GTK+, ubuntu already has things configured and installed.

Ran with this a little. Not so painful.

Geany versions 0.17, 0.18.1, 0.19.2, 0.20 and 0.21 compile cleanly with nekoware gtk2+ 2.12.12 and GCC 4.7.1.
Versions after 0.21 need gtk2+ >= 2.16, which would require me to build a newer gtk2+, which if memory serves won't be an improvement.

NOTE: nekoware libpng12 seems to be incomplete (no static libs). As of geany 0.18 libtool complains about a missing libpng12.al . I replaced nekoware libpng 1.2.4x with a fresh static+shared installation of libpng 1.2.52 - hey presto - no more libtool moaning.

oh, lookey there.... some of Edward Tufte's classic texts on the display of information. Coolness.

I almost bought it. In the end I went with 'Meh'.

Hey all,

Am bumping up against the limits of my (limited) C/C++ knowledge (again). Perhaps somebody can help?

I'm trying to use libcurl from C++. Being pure C libcurl has no idea about C++'s implicit this pointer, so when used from C++ libcurl's callback fn's need to be defined as static . From what I understand, using the static qualifier turns a C++ member variable/function into one 'owned' by the class rather than an instance of that class (or probably something that the compiler sticks outside of the name-mangled mess of all the instances), which in turn removes the need for a this pointer.

So far so good.

My initial solution was to use a few (evil) global variables (why are they evil? they work and don't cause headaches). But when I move those globals into the class definition as static members, I get:


on each of the variables I moved into the class definition. However, the staticly defined class-method callback functions have been working from the beginning.

What am I missing here?

J.

robespierre wrote: to make a function in C++ that uses C calling conventions you normally use the extern "C" {} syntax

Well, the static _functions_ are working just fine. No problems at all. But when I stick the global _variables_ into the private/public sections of the class declaration, that's when I get the ERROR 33's.

foetz wrote: the _state variable of your FTP class is not available to your FTP object file. could you show the FTP class prototypes and one example of using it?

Here you go, the full sources. The issue is in FTP{ .h .C }.

I rolled back the files to the last working versions - ie. the ones without my failed attempts to incorporate the various 'global' variables into the class as static.

Any improvements to the code or tips/comments/suggestions which will improve my coding abilities - the more the merrier

J.

Thank you very very much spiroyster! So nice to get some comments on my code and advice on how to improve it. I'll try your suggestions and see how far I get

If anybody else has other ideas/suggestions, please, don't be shy!

J.

spiroyster wrote: global variables are considered evil in c++, not c (In c there are pretty much an absolute!) for a number of reasons, mainly they imply bad/non-existant oo design.

Took this part of your comments to heart and pulled 'The Gang of Four' off my shelf for the first time since uni (the bill says I bought it in 1997...). Have now re-implemented the FTP class as a Singleton. As you suggested, using a more OO idiom fixed the static/non-static linker thing in one fell swoop. Will refactor the rest of my code to be more OO in design.

Thanks again for taking time to give me your feedback and suggestions. Much appreciated

Awesome punchline

uunix wrote: Not knocking the game, but it does seem comparable with our game of Cricket!

Perhaps you meant to say ' doesn't seem comparable'...

Compared to Baseball, Cricket (in particular Test Cricket) is so much better as a game it's not even a contest

Time to grab a beer and read this:

https://en.wikipedia.org/wiki/Library_%28computing%29

Letterman?

hamei wrote:

yetanother**ixuser wrote:

commodorejohn wrote: Hell, it's basically Lennart Poettering's entire reason for existing...

He-who-must-not-be-named

Jesus. I didn't know anything about all this, did ten minutes' research.

Well, now I guess you understand the context of my rant at viewtopic.php?f=8&t=16727998 a little better...

Go easy on the stout and you'll struggle a good deal less...

That phone was used to read one hamei post too many...

hamei wrote: I'm a little surprised that no one has tried to rescue an xfs disk under Loonix when your last SGI box went kerblooey ?

Really? a whole box going Kerbooey? For me it's never the box itself, it's always a drive. For example, that time when the 9Gb drive in my first O2 died. Which was extra super mega annoying as I had just moved a great deal of data on to it from ZIP-disks and other 'iffy'-media in order to bakkittupp to CD.

For many years I named machines after LotR characters:

*nix main workstations: aragorn, gandalf
*nix one-offs and testbeds: gimli, gloin, bilbo, frodo, faramir
macbook airs: galadriel, legolas
network printer: treebeard
ms windows boxen: troll, orc, gollum, saruman

When I got my HP Proliant microserver,I broke with the Tolkien theme and started naming machines after their external colour:

HP proliant microserver: blackbox
HP zx6000 workstation: silverbox
SGI Fuel: redbox

Hey there Earthlings,

Thrilled to see somebody is looking into using my spit+ducttape Notes thingy. Will look into the various issues and see if/what to fix.

In the meantime here's something completely unrelated yet utterly useful, coz I always forget. Yay!

Hey all,

Can somebody confirm that mediarecorder doesn't provide a way to record a 'screencast' video of the X11 display on the Fuel.

Happy to hear it's my usual braindeadness playing up, and that of course good old SGI made it possible in the first place...

J.

<announcement type="public service">

Some of you lot might have had to do with Real Colour on Real Printing Machines. Some of you might not.

Either way, this is your friendly reminder that colour calibration can be a bit of a Black Art and that most OpenSource gfx software still has issues when it comes to colour for print. Not entirely the OpenSource community's fault, but nonetheless the Scanner->GIMP->Inkscape->PDF->Real Printing Machinery workflow isn't without speedbumps.

</announcement>

And because the interweb provideth:

http://www.thebookdesigner.com/2010/03/ ... -designer/

Enjoy

The opensource toolset/chain is well-suited to on-screen and non colour critical work. it's just a bit of a mess when you need colour accuracy.

@Hamei - when I get home from the Carnival I'll give Scribus and Inkscape a go see if I can get something 'worthwhile' out of them.

@Hamei: Have given up on getting scribus to work. The neko packages from current dont run on my system and the new scribus needs versions of qt that wont compile on our beloved 20 yr old machines. Oh well.

I'm messing with the hardware configuration of redbox and I've noticed that the PSU crackles a bit when I insert the power cable and that I can only start the machine after the crackling settles down.

As I know very little about hardware, I'm assuming this is normal operation and that the crackling is caused by (maybe) the capacitators 'filling up' with electricity?

Well, maybe it's the pre-death clicking then... after I insert the cable into the PSU, for about 2-3 minutes there's a very soft 'crackling' to be heard coming from the PSU. During this time the PSU does not switch on. After a while the crackling subsides and then the PSU switches on without any issue.

It only happens after the PSU has been disconnected from the wall socket. Cycling/start/stopping the machine without removing the lead from the wall doesn't affect the machine.

In my Make-The-Fuel-Quieter project of a while back I not only replaced the fans and got some good results, but I also used some of the silicone washers I got with the fans to dampen parts of the casing. In particular, i shoved washers down one side of each vrpo fan to stop them from rattling in their blue shroud.



Which got me thinking: redbox stil has some annoying sound characteristics. Perhaps there's more I can do.

So, this morning I opened it up and did some prodding and poking and discovered that the system drive mounted in the 3.5" middle drive bays with the blue SGI-supplied drive rails was vibrating the 5.25" drive bays above it, resulting in an audible drone. Prodding/poking/dampening the bottom drive bay fan also resulted in a reduction of drone.



Armed with the new prodding data, I hopped out to Maplins and got a packet of rubber fan mounts.



Cutting off some of the fan mount rubber allowed me to remount the middle drive bay fan using the original (long-ish) screws with the cut-down rubber mounts on the diagonal. (4 fan mount holes, two screws, two rubber mounts). The result is a significant reduction in resonance drone eminating from that fan and the surrounding case metal.

I then removed the system drive from the middle drive bays and booted redbox - obviously the resonance drone from the top 5.25" bays disappeared completely. However, when placed in the bottom bays the drive produced a new resonance, but padding the bottom bay with silicone washers has significantly reduced the new resonance.

Top tip: make a long masking tape inside out sticky thing first, then place the washers on the sticky thing and cut between each washer to reduce clumsy finger tape stickage when trying to make perfect 7mm inside out masking tape loops.



The drive just sits in the bay, no screws or fixings are used.



Given that I plan to re-install IRIX, I proceeded to test each of the SCSI/SCA drives I have to find the quietest one. It seems that my 147Gb 15k drives all produce a piercing high-pitched whine. Some of the 10k units of various sizes just made a lot of noise in general. I've settled on a 36Gb 10k Fujitsu unit re-badged by HP. It too has a whine, albeit at a softer volume and most importantly at a lower and more sufferable pitch. I do not have enough drives at hand to offer recommendation for a model/make/type, but it seems fair to say that 15k units will always emit a very high pitched whine.



Finally, padding the ends of the case crossbar (what a POS the Fuel case is) with some of the silicone washers seem to have removed some noise/vibration too.

But at this point I might be starting to hear things

Pontus wrote: Nice work Which fans did you replace and what did you replace them with?

Specs and story at the bottom of this old thread:

viewtopic.php?f=3&t=16725732

GL1zdA wrote: Does the Scythe Kama Flow in your PSU require suspending environmental monitoring?

'require' - no it didn't. It worked just fine. Past tense. A couple of months after replacing the original fan, that PSU failed. I lack the hardware skills and tools to do any forensics so it's hard to say if it was the fan or if it just failed from 'failure'. So, I blamed myself for being adventurous, got a new PSU and to be on the safe side didn't replace the fan again. Since then redbox has seen some resonably long working hours without any issues.

But now I have a spare PSU on the shelf and maybe it's time to be brave again. If the PSU dies again we can be rpetty sure it was the fan replacement.

In fact, if I still have a spare quiet fan lying about that's exactly what I'm going to do... watch this space.