blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates AAD/TAB MS361 x42854") (07/26/90)
The last I heard was that the IBM's were comparable to the bottom of the SGI line, 4D/20 with minimal graphics. -- Brent L. Bates NASA-Langley Research Center M.S. 361 Hampton, Virginia 23665-5225 (804) 864-2854 E-mail: blbates@aero4.larc.nasa.gov or blbates@aero2.larc.nasa.gov
dhinds@portia.Stanford.EDU (David Hinds) (07/29/90)
In article <9007261139.AA05802@aero4.larc.nasa.gov> blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates AAD/TAB MS361 x42854") writes: > > The last I heard was that the IBM's were comparable to the bottom >of the SGI line, 4D/20 with minimal graphics. >-- The IBM RS6000's have several levels of graphics support. The 8-bit color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D polygons/sec. The 24-bit color 3D graphics system is quoted as doing 990K 3D vectors/sec and 120K 3D polygons/sec. For comparison, SGI says a base IRIS 4D/50 with 8 bit planes does 140K vectors/sec and 5.5K polygons/sec. An IRIS with GTX graphics is supposed to do 475K vectors/sec and 100K polygons/sec. It is not obvious how to compare the figures, however. The IBM report quotes the lengths of vectors and sizes of polygons used. I looked through all our SGI stuff and couldn't find the corresponding information. These were the same SGI tables that quote 100 MIPS and 50 MFLOPS for a 4D/240, which are the theoretical limits, rather than performance on any standard benchmark. -David Hinds dhinds@popserver.stanford.edu
ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) (07/31/90)
In article <1990Jul29.165033.22289@portia.Stanford.EDU>, dhinds@portia.Stanford.EDU (David Hinds) writes: > In article <9007261139.AA05802@aero4.larc.nasa.gov> blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates AAD/TAB MS361 x42854") writes: > > > > The last I heard was that the IBM's were comparable to the bottom > >of the SGI line, 4D/20 with minimal graphics. > >-- > The IBM RS6000's have several levels of graphics support. The 8-bit > color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D > polygons/sec. The 24-bit color 3D graphics system is quoted as doing > 990K 3D vectors/sec and 120K 3D polygons/sec. Has it been released yet or is it still vaporware? Something else to consider is that IBM's high-end graphics board is an IBM proprietary and does not run the GL. > For comparison, SGI says a base IRIS 4D/50 with 8 bit planes does > 140K vectors/sec and 5.5K polygons/sec. An IRIS with GTX graphics is > supposed to do 475K vectors/sec and 100K polygons/sec. Don't forget the VGX system's 1M vps and 1M pps. This board uses the GL unlike IBM's top-o'-the-line. > It is not obvious how to compare the figures, however. The IBM > report quotes the lengths of vectors and sizes of polygons used. I looked > through all our SGI stuff and couldn't find the corresponding information. > These were the same SGI tables that quote 100 MIPS and 50 MFLOPS for a > 4D/240, which are the theoretical limits, rather than performance on any > standard benchmark. I'm not sure what tables you are looking at but the tables I've seen SGI quote for the 4D/240 are 4 cpus x 20 VAX Dhrystone MIPS = 80 MIPS and a floating point performance 16MFLOPS double-precision Linpack. I've never heard they numbers you quote. The graphics performance numbers for the Personal Iris and GT graphics systems vps are based on 10 pixel, connected, full 24-bit color, arbitrary orientation vectors. For pps, 10x10 (100 pixel), full 24-bit color, unlighted, Gouraud shaded, Z-buffered, arbitrary orientation, polygons. The numbers for the VGX system are based on anti-aliased vectors and polygons in triangle meshes. > > -David Hinds > dhinds@popserver.stanford.edu My own personal opinion is that most quoted benchmarks are useful only as a guideline, actually mileage may vary due to driving conditions. The really interesting performance comparisons come for independendant vendors who run application scenario comparisons between various platforms. These aren't benchmarks in the traditional sense of having a very specific test of floating point performance. These comparisons are based on tests that similate trying to do actual work and take into account full system performance (CPU, disk subsystem, graphics, et cetera.) These, to me, represent the really interesting comparison points. Some of the vendors use these numbers to determine the pricing of their software on various platforms so it would seem these kinds of comparisons are interesting to them too. Of course, the bottom line is what is the real performance of your personal tasks. If you can port your application to make the comparison. --- Ciemo
dhinds@portia.Stanford.EDU (David Hinds) (07/31/90)
In article <11111@odin.corp.sgi.com> ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) writes: >In article <1990Jul29.165033.22289@portia.Stanford.EDU>, >dhinds@portia.Stanford.EDU (David Hinds) writes: >> The IBM RS6000's have several levels of graphics support. The 8-bit >> color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D >> polygons/sec. The 24-bit color 3D graphics system is quoted as doing >> 990K 3D vectors/sec and 120K 3D polygons/sec. > >Has it been released yet or is it still vaporware? Something else to >consider is that IBM's high-end graphics board is an IBM proprietary >and does not run the GL. Is this right? I thought that all the 3D graphics options for the IBM's were SGI technology. An IBM rep told me they would support GL. I don't know if it has been shipped yet. >> For comparison, SGI says a base IRIS 4D/50 with 8 bit planes does >> 140K vectors/sec and 5.5K polygons/sec. An IRIS with GTX graphics is >> supposed to do 475K vectors/sec and 100K polygons/sec. > >Don't forget the VGX system's 1M vps and 1M pps. This board uses the GL >unlike IBM's top-o'-the-line. Yeah, my table didn't include VGX. >The graphics performance numbers for the Personal Iris and GT graphics systems >vps are based on 10 pixel, connected, full 24-bit color, arbitrary orientation >vectors. For pps, 10x10 (100 pixel), full 24-bit color, unlighted, Gouraud >shaded, Z-buffered, arbitrary orientation, polygons. The numbers for the VGX >system are based on anti-aliased vectors and polygons in triangle meshes OK. The IBM vectors are the same as the SGI ones. The sheet says the polygons were 50 pixel Gouraud shaded triangle mesh. I don't know how the speed scales with size. -David Hinds dhinds@popserver.stanford.edu
shenkin@cunixf.cc.columbia.edu (Peter S. Shenkin) (07/31/90)
In article <1990Jul30.210557.6255@portia.Stanford.EDU> dhinds@portia.Stanford.EDU (David Hinds) writes: >In article <11111@odin.corp.sgi.com> ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) writes: >>In article <1990Jul29.165033.22289@portia.Stanford.EDU>, >>dhinds@portia.Stanford.EDU (David Hinds) writes: >>> The IBM RS6000's have several levels of graphics support. The 8-bit >>> color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D >>> polygons/sec. The 24-bit color 3D graphics system is quoted as doing >>> 990K 3D vectors/sec and 120K 3D polygons/sec. >> >>Has it been released yet or is it still vaporware? Something else to >>consider is that IBM's high-end graphics board is an IBM proprietary >>and does not run the GL. > > Is this right? I thought that all the 3D graphics options for the IBM's >were SGI technology. An IBM rep told me they would support GL. I don't >know if it has been shipped yet. It's definitely true that IBM's high-end, 990K-vector/sec graphics board set is IBM-proprietary. They claim that it can do certain things that the GL library does not support, namely the entire PHIGS model. (I don't even know exactly what this means, so don't flame me if this seems not to make sense. I'm just repeating what I was told by an IBM rep!) But more recently I've seen literature that seems to imply that IBM has a software library of GL calls which have been ported to these boards as well. The SGI rep told me that the IBM low-end graphics board (90k 3D vectors/sec) is the basic 4D/20 set which SGI has licensed to IBM, and that this is the only hardware that SGI has licensed to IBM. As computational engines, the IBM machines seem superb, especially the Model 540, which is designed as a number-cruncher, with very fast clock and memory, and is not available with fancy graphics. I have been told second-hand that people are getting Cray-1 speeds for vectorizable floating-point code out of this workstation. The graphics models also seem very fast, and all are priced very competitively. It seems to me that there aren't enough of these machines out there yet to know how they really stack up, though; until then, you gotta decide if you want to take a risk with RISC-6000. And SGI's strong software support base, as much as anything else, has made me decide to go with SGI for my upcoming purchase. -P. ************************f*u*cn*rd*ths*u*cn*gt*a*gd*jb************************** Peter S. Shenkin, Department of Chemistry, Barnard College, New York, NY 10027 (212)854-1418 shenkin@cunixc.cc.columbia.edu(Internet) shenkin@cunixc(Bitnet) ***"In scenic New York... where the third world is only a subway ride away."***
sergio@sergio.uucp (Sergio Perrone/30000) (07/31/90)
The RS/6000 Model 540 can handle the SGI-compatible graphics board... Marc --- Marc Andreessen, IBM AWD Austin, sergio@sergio.austin.ibm.com --- --- Words and ideas contained herein are independent of IBM policy. ---
ra@sun1.ruf.uni-freiburg.de (Christoph Ramshorn) (07/31/90)
dhinds@portia.Stanford.EDU (David Hinds) writes: > The IBM RS6000's have several levels of graphics support. The 8-bit >color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D >polygons/sec. This level of perfomance comes also with 24-bit color and 24-bit z-buffer. It is based on SGI chips and runs under GL - comparable to a 4D20/25. It is > The 24-bit color 3D graphics system is quoted as doing >990K 3D vectors/sec and 120K 3D polygons/sec. This is the graphics of the 730 model. It is IBM hardware and does NOT run with GL. Christoph Ramshorn, Geologisches Institut, Albertstr.23-B D-7800 Freiburg i.Br., West Germany
archer@elysium.esd.sgi.com (Archer Sully) (07/31/90)
In <1990Jul30.210557.6255@portia.Stanford.EDU> dhinds@portia.Stanford.EDU (David Hinds) writes: *In article <11111@odin.corp.sgi.com> ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) writes: *>In article <1990Jul29.165033.22289@portia.Stanford.EDU>, *>dhinds@portia.Stanford.EDU (David Hinds) writes: *>> The IBM RS6000's have several levels of graphics support. The 8-bit *>> color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D *>> polygons/sec. The 24-bit color 3D graphics system is quoted as doing *>> 990K 3D vectors/sec and 120K 3D polygons/sec. *> *>Has it been released yet or is it still vaporware? Something else to *>consider is that IBM's high-end graphics board is an IBM proprietary *>and does not run the GL. * * Is this right? I thought that all the 3D graphics options for the IBM's *were SGI technology. An IBM rep told me they would support GL. I don't *know if it has been shipped yet. IBM has three levels. The base model has no graphics accellerator. The mid-range has the Personal Iris (GR1) graphics squeezed onto MicroChannel(TM) cards. Their high-end system (the one no one that I know has seen) was developed at IBM, and beyond that I know nothing about it. It is quite possible that they've ported the GL to run their high-end system, but that means that they'll only be tracking what we've sold them to date. Which says to me that they won't be offering full compatibility with the current incarnation of the GL. As usual I don't speak for SGI or IBM. -- Archer Sully | Ask not what you can do for you country archer@esd.sgi.com | But what your country's been doing to you. | -- The Avengers
kuba@hpfcdj.HP.COM (Kathy Kuba) (08/22/90)
>*> >*>Has it been released yet or is it still vaporware? Something else to >*>consider is that IBM's high-end graphics board is an IBM proprietary >*>and does not run the GL. >* >* Is this right? I thought that all the 3D graphics options for the IBM's >*were SGI technology. An IBM rep told me they would support GL. I don't >*know if it has been shipped yet. > > >It is quite possible that they've ported the GL to run their high-end >system, but that means that they'll only be tracking what we've sold them >to date. Which says to me that they won't be offering full compatibility >with the current incarnation of the GL. At SIGGRAPH, IBM told me that GL runs across their entire 3D product line, which includes the Personal Iris boards on a 320/530 and the 730 with proprietary supergraphics. It is supposed to be code compatible with SGI, but I do not know of anyone who has tried this yet--anyone out there know for sure?? Their PHIGS (GraPHIGS) also runs across the entire line. Kathy
mla34142@uxa.cso.uiuc.edu (08/24/90)
/* Written 4:06 pm Aug 21, 1990 by kuba@hpfcdj.HP.COM in uxa.cso.uiuc.edu:comp.sys.sgi */ /* ---------- "Re: New IBM Graphics Workstations" ---------- */ >At SIGGRAPH, IBM told me that GL runs across their entire 3D product line, >which includes the Personal Iris boards on a 320/530 and the 730 with >proprietary supergraphics. It is supposed to be code compatible with >SGI, but I do not know of anyone who has tried this yet--anyone out >there know for sure?? Their PHIGS (GraPHIGS) also runs across the entire >line. > >Kathy GL will run across the entire product line, including the 730. It is code compatible, with certain exceptions, mostly related to the fact that IBM GL runs through X (as opposed to SGI's, which uses NeWS). For example, the mapcolor() call shouldn't be used because mapcolors() can do colormap changes 4096 times as fast. An additional header file, gl/glport.h, is also included to help in porting programs with obsolete function calls, and can be used to ease the porting process. Marc
kuba@hpfcdj.HP.COM (Kathy Kuba) (09/05/90)
> / hpfcdj:comp.sys.sgi / ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) / 12:10 pm Jul 30, 1990 / > In article <1990Jul29.165033.22289@portia.Stanford.EDU>, > dhinds@portia.Stanford.EDU (David Hinds) writes: > > In article <9007261139.AA05802@aero4.larc.nasa.gov> > blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates AAD/TAB MS361 x42854") writes: > > > > > > The last I heard was that the IBM's were comparable to the bottom > > >of the SGI line, 4D/20 with minimal graphics. > > >-- > > The IBM RS6000's have several levels of graphics support. The 8-bit > > color 3D graphics level is quoted as doing 90K 3D vectors/sec, and 10K 3D > > polygons/sec. The 24-bit color 3D graphics system is quoted as doing > > 990K 3D vectors/sec and 120K 3D polygons/sec. > > Has it been released yet or is it still vaporware? Something else to > consider is that IBM's high-end graphics board is an IBM proprietary > and does not run the GL. > An IBM rep at SIGGRAPH told me that GL and GraPHIGS both run across their entire 3D line, which includes the PI boards. If you have contradictory information, I am very interested in learning more. > > > For comparison, SGI says a base IRIS 4D/50 with 8 bit planes does > > 140K vectors/sec and 5.5K polygons/sec. An IRIS with GTX graphics is > > supposed to do 475K vectors/sec and 100K polygons/sec. > > Don't forget the VGX system's 1M vps and 1M pps. This board uses the GL > unlike IBM's top-o'-the-line. > > > It is not obvious how to compare the figures, however. The IBM > The graphics performance numbers for the Personal Iris and GT graphics systems > vps are based on 10 pixel, connected, full 24-bit color, arbitrary orientation > 2D or 3D vectors? Kathy
robert@texas.esd.sgi.com (Robert Skinner) (09/06/90)
In article <17450006@hpfcdj.HP.COM>, kuba@hpfcdj.HP.COM (Kathy Kuba) writes: |> > / hpfcdj:comp.sys.sgi / ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) / 12:10 pm Jul 30, 1990 / |> > |> > > For comparison, SGI says a base IRIS 4D/50 with 8 bit planes does |> > > 140K vectors/sec and 5.5K polygons/sec. An IRIS with GTX graphics is |> > > supposed to do 475K vectors/sec and 100K polygons/sec. |> > |> |> 2D or 3D vectors? |> |> Kathy unlike some other companies nearby, SGI's benchmarks are *always* for *fully transformed and clipped 3D* vectors. Robert Skinner robert@sgi.com Fish, he got a hook in his throat. Fish, he got problems. - Genesis
ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) (09/06/90)
In article <17450006@hpfcdj.HP.COM>, kuba@hpfcdj.HP.COM (Kathy Kuba) writes: > > / hpfcdj:comp.sys.sgi / ciemo@bananaPC.wpd.sgi.com (Dave Ciemiewicz) / 12:10 pm Jul 30, 1990 / > > > > > For comparison, SGI says a base IRIS 4D/50 with 8 bit planes does > > > 140K vectors/sec and 5.5K polygons/sec. An IRIS with GTX graphics is > > > supposed to do 475K vectors/sec and 100K polygons/sec. > > > > Don't forget the VGX system's 1M vps and 1M pps. This board uses the GL > > unlike IBM's top-o'-the-line. > > > > > It is not obvious how to compare the figures, however. The IBM > > The graphics performance numbers for the Personal Iris and GT graphics systems > > vps are based on 10 pixel, connected, full 24-bit color, arbitrary orientation > > > > 2D or 3D vectors? > > Kathy 3D, of course. Sorry for the missing detail. --- Ciemo