The collected works of robespierre - Page 1

josehill wrote:
squeen wrote:
- 1st touch screen on a phone
Not quite - Palm Treos had touch screens several years before iPhones came out, though obviously the iPhone implementation was more sophisticated.


I believe squeen was referring to this Apple concept phone from 1983. Since Jobs was on the Macintosh project at the time, he probably can't take credit for it.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
When it rains, it pours. John McCarthy died today .
I suppose the MSM coverage will be even sparser yet.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I remember that Nendo from Nichimen (a polygon modeler for games and simulations) was released on Solaris. There was also a period when some companies were making digital video capture and editing products for Sparc, and Evans and Sutherland sold 3D graphics hardware for them.

Also don't forget HyperLook , a NeWS framework that was used for some commercial apps.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I think Electric Image should say EI Universe, as that was the only version that was ported to Unix.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
oreissig wrote: although that doesn't always mean a thing, I also installed stock IRIX 5.3 on my Indigo² 200MHz R4400 1MB, although it only supports 100 MHz R4000SC or 150 MHz R4400SC (according to sgistuff )
and the 250MHz 2MB CPU is not supported by IRIX 5.3, but good old 4.0.5H runs fine on it :D


It is a different story with the 2MB SC processor, first 5.3 release definitely does not work.
And 250MHz R4400 Indigo2 is indeed supported by IRIX 5.3 with appropriate patches. (SG0951,956,1003,1015).

To the OP: J.J. identified the reason for your problem, but I would add that [A]uto, or INITIALIZE command to fx is not recommended on IRIX 5.3, because due to a bug it performs a low-level format. This was fixed for IRIX 6.2 fx.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
Most printed circuits manufactured after 2000 use SnAgCu solder alloys. Whatever the melting point, it can be difficult to reflow joints attached to ground planes.

I recently discovered a tool that simply cuts off chip capacitors, it might work on the smaller OSCONs as well.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
man cc:

* The gcc compiler allows variadic macros; the MIPSpro 7.4 compilers
support these macros in c99 mode. If you have code that uses
ellipses (...) as part of a macro definition and you are not
compiling with c99, you will need to rewrite the macro. Two possible
approaches are to replace the macro with a new variadic function, or
to create a family of macros, each taking different (fixed) numbers
of arguments.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
from kubatyszko's remarks, and the Sanyo datasheets, these are surface-mount electrolytic caps. they need SMD techniques.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
This may be a naive suggestion, but couldn't you use a C99 cpp pass using -P and run CC on that?

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
yes, the presenter board has a minidin-3 stereo connector. But the stereo sync signal may be available from another place within the IP32 itself.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
as long as you choose a sysid that is a valid unicast MAC address, the onboard ethernet should work with any address you want. the reason for keeping the same address is for software licensing, and to avoid confusing routers/switches that have the station MAC address stored, for VLAN assignment, proxy ARP, etc.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
guardian452 wrote: nail, head, hit. I've disabled the on-screen clock on my computers. It was (rather surprisingly) very scary at first. I suggest everybody try it as an experiment ;) You might be surprised how often you subconsciously are checking the time...


See J.G. Ballard's story "Chronopolis" in this connexion.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
Yes, the stereo sync port is for use with the VGA output, because neither the Presenter 1280 panel nor the 1600SW panel is capable of stereo refresh rates. You either use the flat panel port or the stereo port, not both at once.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
those Cipher tape units are neat, I've always been impressed by half-inch tape machines.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
Perhaps the poster meant Alias Studio Tools?

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
Also there was no XS graphics for Indigo2, there was only XZ (soon updated to 4 GE or Elan equivalent) and Extreme (8 GEs) within the GR3 family.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
noticed that the O2 docs mention a 256MB limit for direct kernel access to memory.
Code:
For system memory, the full 1GByte memory space is directly accessible to the CPU, but the first 256 MBytes are aliased to KSEG0 so that operating system structures may be mapped without using TLB entries. To access the full 1GByte memory configuration, translation buffer entries are required.


The R5000 has only 48 TLB entries, so there may be more pressure on this resource when capturing in a >128 MB system.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
The R10K has more TLB entries (64) but as far as I know the memory map works the same way.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
24-bit color (8 bits per component) has been in use since the 1970s. As has raytracing, texture mapping, Gouraud and Phong shading, etc. Even radiosity is from the early 1980s. Now obviously rendering technology has advanced considerably, but for many of the application areas you mentioned, technology like PRMan is still some kind of standard for rendering, and it is from the late '80s.

I think the cause of the different "look" is that computers were much less powerful, so scenes were much simpler, and it was still fashionable to have very shiny, brightly colored characters. Part of this is the space-age generation: the visual look of something like Tron owes not a little to Logan's Run or THX 1138. Part of it is again the limits of computer power, and typically very small textures by today's standards. There were also much more primitive software tools, so while the renderer may have been capable of a richer and more modern-looking scene, it would have taken too many man-years to design it. Keep in mind that before Photoshop and its peers making a texture was incredibly complicated.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
it depends on your definition of workstation. The IRIS 2000 was released in 1984 or 1985, and Symbolics had something at around the same time. Both systems had 3D modeling software. There was also Quantel Paintbox and other stuff but I'm not sure that counts in terms of 3D animation.

Yes, environment mapping was widely used in the 80s, it is a natural application of texture mapping. There were many schemes for interpolating textures in either world space or screen space and a lot of hardware was designed at the time. The use of primitives is especially seen in very early animations (see for example "André and Wally B.") but I think by the late '80s there were many more organic characters. NURBS were developed to make modeling organic shapes easier, don't know if I would call that a success. ;)

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
(the fairly realistic trees in that short are actually decals/impostors I think. still they look quite effective in context.)

The very first film-quality 3D graphics was all done on custom, one of a kind hardware. You can see pictures of some of it here:
http://dave.zfx.com/omnibus.html
Important to note, I think, is that those systems were not optimized around real-time display of models but pure rendering and high-resolution output ability. The scenes would have been developed using vector display terminals like tektronix 4014 or other similar things, almost entirely as wireframes.

Much early TV animation (for station IDs etc) was actually done on analog video synthesizers that had no 3D software, or really any software. See pictures here . They were often able to trick the eye and you may not have realized the difference. Analog animations were all done in real-time direct output to videotape, whereas the Triple-I/Omnibus productions were all scanned frame-by-frame directly to 35 or 70 mm film.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I think that Geometry Engine type graphics pipelines only supported what X calls DirectColor schemes. So 3:3:2, 5:5:5, 6:6:6, 8:8:8, or 10:10:10 would have been possible. The IRIS 2000 could be configured with three BP3 bitplane boards, and there are man pages on the Web that describe "the 24 bits of color available on the IRIS 2400/3000".

The first Symbolics workstation capable of 24-bit color (although I think 32 was more common) was the 3600, released in 1984.

hamei: you may be right, but I first learned of NURBS because Wavefront was promoting them.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
There is another listing for what appears to be the same type of unit, at
http://www.ebay.com/itm/SGI-Silicon-Gra ... 0551118018

I would guess an industrial automation product, maybe for optical recognition of parts/defects in a manufacturing plant. No way to really know unless somebody buys one and looks at the boards.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
hamei, while you're right that everyone's a crook :roll: ebay has had the "mouse over image to zoom" feature for years. Until a few days ago you needed to click "zoom" to activate it, but now they have made it automatic.

Maybe you will enjoy this link if you have not seen it already:
http://www.sanskritweb.net/forgers/

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
My understanding is that fonts are a grey area. The rule that distinguishes works that can be copyrighted from those that may be patented is that the former have some creative, and not useful, part. It may be said that while works of art or literature may have an effect on the reader, they are not really tools used to create effects, but rather _expressions_ of the author or artist. A typeface is a tool used to create an effect, namely legibility when arranged on a printed page with other elements, and is not usually an expression of thoughts or feelings in itself. Fancy display fonts sometimes blur the line however.

The modern font file (which Adobe likes to call a font "program") is _more than_ the old utilitarian typeface, because it contains parts that have no particular purpose and are merely expressions. For example, the comments or the order in which the points are drawn. :roll: So they have copyrights attached and the companies (which still style themselves "foundries") are very protective of them.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
gmemusage and gr_osview are the most obvious.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I thought that the R10K/R12K set its clock frequency (PClk, a multiple of the SysClk signal from the board) based on a 64-bit word written by the chipset at reset time. So I don't understand how moving resistors can affect that. The multiplier is part of the data contained in the serial ROM on the module, I thought.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
easier said than done I'm sure. There was a screensaver that mimicked ElectroPaint done several years ago, but it isn't even close.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
recondas, in the two PDFs you linked, the 3M one and the CameraLink one, the "pinouts" or connectivity between the plugs is the same. Both are simple rollover configurations (which is actually a simpler configuration than "straight thru" for connectors with multiple rows). But the internal construction of the cables is different, as one has 11 unshielded twisted pairs and the other has 6 twinax shielded cables plus 8 single wires. This difference can't be distinguished by simply probing with a continuity tester. So suppose that the cable SGI uses tests as the same pinout as either of these. That wouldn't be sufficient information to know that it would work as a replacement: the characteristic impedance of the signals may be different.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
That is the "Indigo2 Video for Impact" board. http://lurkertech.com/lg/ev1s.html
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I'm not an expert on importing/exporting, but I would assume that the parcel is sealed inside a cargo container for most of that time. Which doesn't necessarily make it safe...

Image

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I had the opportunity to buy a Polaroid slide recorder once, but I wasn't sure if it could be adapted to 4-perf cine format.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I never had a MV2000 specifically, but with many DEC desktops of that era the sheet metal cover is held on with a lot of friction (stiff metal EMI springs). Try holding the sides while letting the chassis drop out the back.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
You have a PM.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
yeah I've seen those at SIGGRAPH. a lot of data/screen area is wasted for views facing where there are no people—it makes the most sense for public displays and installations.

More impressive are multilayer displays—I think there was one with a thousand layers. Very heavy horsepower required for realtime displays...

here's a cool link: http://web.media.mit.edu/~dlanman/resea ... index.html
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
the specific reason is that, except when used as the argument to operator& or sizeof(), array expressions are converted to a pointer to their first element.
This is handy for C's function calling rules, which do not allow array types to be passed to, or returned from functions.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
No, you need HPUX 11i v1 (11.11).
See http://www.openpa.net/hp-ux_unix.html#hpux11iv1
11.0 doesn't support PA8800/8900 processors.

The license is transferrable with the hardware, so you should be able to buy a used media set.

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
I found a book where Buñuel quotes Wisdom 2:1-9 out of context, and it impressed me a lot.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
fu wrote: try to find "the exterminating angel"


That's a great movie. Buñuel said of Mexico, "A fascist country softened by corruption."

I'll try to see it when the local place shows it in August.
:PI: :O2: :Indigo2IMP: :Indigo2IMP:
uridium wrote:
The port to alpha shows pretty clearly you don't need exotic priv levels in your hardware any more.


How do you figure? The Alpha is not what I'd call "exotic", but it is significantly richer from an OS support standpoint than its erstwhile competitors.

Code:
VAX features in the Alpha AXP architecture that are important to OpenVMS system software are: four protection modes, per-page protection, and 32 interrupt priority levels.

[Digital Technical Journal Vol. 4 no. 4, §7]

_________________
:PI: :O2: :Indigo2IMP: :Indigo2IMP: