Diamond EDGE or 3D BLASTER... which will be better??
Hamid R. Moazed
> From: td3...@u.cc.utah.edu (Travis Driscoll)
> Newsgroups: comp.sys.ibm.pc.games.action,comp.sys.ibm.pc.hardware.video,comp.sys.ibm.pc.soundcard.tech
> Followup-To: comp.sys.ibm.pc.games.action,comp.sys.ibm.pc.hardware.video,comp.sys.ibm.pc.soundcard.tech
> Date: 26 Sep 1995 17:08:52 GMT
> Organization: University of Utah Computer Center
>
> dani...@vnet.ibm.com wrote:
> : In <444f87$p...@news1.wolfe.net>, fir...@wolfenet.com (Dan Osborn) writes:
> : >>For one thing, the Edge's sound abilities should blow the AWE out of the
> : >>water. I've got an AWE, I should know. It is (the AWE), however, a nice
> : >>composing card. Most likely you won't get 100% SB compatibility, but what
> : >>non-CL can really claim that. Advantages of the Edge's sound, hardware
> : >>FX mixing (can be interpolated to 16bit, 44.1KHz), and Fat Labs certified
> : >>MIDI music with either 1MB or 6MB sound bank. Fat Labs (a group of
> : >>composers responsible for music from titles like Nascar, WC, Loom, 7th Guest,
> : >>Seawolf, etc...) compares cards to Roland Sound Canvas. CL is still
> : >>undergoing this certification and no hardware mixed FX are implemented.
> : >
> : >Fred, what about this from diamond release:
> : >
> : >"Wavetable Audio
> : >
> : >A high-fidelity hardware wavetable audio engine is integrated into the
> : >Diamond Edge 3D. The wavetable samples are stored in system memory
> : >allowing
> : >game developers to provide custom samples for exciting sound and MIDI
> : >effects tailored to each individual game. Efficiently designed to
> : >share
> : >system resources and reduce total system cost, Edge 3D uses the
> : >computer's
> : >main memory to hold wavetable samples. Unlike most wavetable products
> : >available today, which require expensive additional read only memory,
> : >new
> : >sound samples can simply be downloaded for use with the Diamond Edge
> : >3D."
>
> : So does this mean the Edge could be used as a MIDI sampler? The
> : PCI bus should be fast enough, certainly faster than loading
> : patches to an Ultrasound from hard disk. Plus, instead of
> : loading up the soundcard with 8 megs of ram, you could just
> : put that ram in your main system. That way it can be used
> : for other things, instead of rotting in the soundcard all day.
> : If Diamond makes this thing MIDI compatible with patch
> : caching applications like Cakewalk it could be useful for more
> : than games.
>
> Yikes! Anytime something on the bus tries to access main RAM it has to
> steal cycles from the CPU!
> The CPU has to sit on it's hands waiting for memory as it is...
> Expect a serious performance drop if your game is playing intesive
> midi.
> And anyone who uses main memory for video will shut out the CPU whenever the
> screen has to refresh. Why do you think a VRAM card is faster? During
> refresh, VRAM can be read from by the DAC as well as accessed by the
> system at the same time. With DRAM, if the system tries to address
> an area of memory that's being accessed by the DAC refreshing the screen,
> the system is put in a wait state until that section of memory is free.
> These restrictions are only placed on video memory on the card, but
> imagine if the DAC were stepping on toes in main memory?
>
> I think it's a terrible idea.
> --
> Travis Driscoll
> td3...@u.cc.utah.edu
> This space intentionally left blank.
With a 66MHz 32-bit memory bus, reading 44kHz 16-bit samples (CD-quality) from memory will use up 0.032% of the memory's bandwith. I don't think anyone can feel a performance penalty of .00032, and even on a 33MHz 16-bit bus it is still imperceptable at just above .1%. Your argument about Video is true, but the bandwith of video is orders of magnitude above that of audio (640x480x16 at 60Hz requires nearly 37 million bytes per second, while CD-quality audio needs 88 thousand bytes per second, about 400 tim
es less. Unless you have a 4.88MHz 8088 you shouldn't feel much performance penalty for audio DMA.
-Hamid
Tow Wang
Hamid R. Moazed <ha...@credo.west.rs.com> wrote:
>per second, while CD-quality audio needs 88 thousand bytes per second,
88200 samples (not bytes) per second.
Tow
-----------------------------------------------------------------------------
The FAQ file 1.04 for the TB Tropez sound board can be obtained by FTP from:
ftp.wi.leidenuniv.nl /pub/audio/tropez/
ftp.hawaii.edu /outgoing/Maui/
ftp.netcom.com /pub/jo/jonin/soundcards/faqs/
-----------------------------------------------------------------------------
Walter Meyer
>In article <449c34$p...@news.cc.utah.edu> td3...@u.cc.utah.edu (Travis Driscoll) writes:
>> From: td3...@u.cc.utah.edu (Travis Driscoll)
>> Newsgroups: comp.sys.ibm.pc.games.action,comp.sys.ibm.pc.hardware.video,comp.sys.ibm.pc.soundcard.tech
>> Followup-To: comp.sys.ibm.pc.games.action,comp.sys.ibm.pc.hardware.video,comp.sys.ibm.pc.soundcard.tech
>> Date: 26 Sep 1995 17:08:52 GMT
>> Organization: University of Utah Computer Center
>>
>> dani...@vnet.ibm.com wrote:
(Stuff deleted)
>> Yikes! Anytime something on the bus tries to access main RAM it has to
>> steal cycles from the CPU!
>> The CPU has to sit on it's hands waiting for memory as it is...
>> Expect a serious performance drop if your game is playing intesive
>> midi.
>> And anyone who uses main memory for video will shut out the CPU whenever the
>> screen has to refresh. Why do you think a VRAM card is faster? During
>> refresh, VRAM can be read from by the DAC as well as accessed by the
>> system at the same time. With DRAM, if the system tries to address
>> an area of memory that's being accessed by the DAC refreshing the screen,
>> the system is put in a wait state until that section of memory is free.
>> These restrictions are only placed on video memory on the card, but
>> imagine if the DAC were stepping on toes in main memory?
>>
>> I think it's a terrible idea.
>> --
>> Travis Driscoll
>> td3...@u.cc.utah.edu
>> This space intentionally left blank.
>With a 66MHz 32-bit memory bus, reading 44kHz 16-bit samples (CD-quality) from memory will use up 0.032% of the memory's bandwith. I don't think anyone can feel a performance penalty of .00032, and even on a 33MHz 16-bit bus it is still imperceptable at just above .1%. Your argument about Video is true, but the bandwith of video is orders of magnitude above that of audio (640x480x16 at 60Hz requires nearly 37 million bytes per second, while CD-quality audio needs 88 thousand bytes per second, about 400 tim
>es less. Unless you have a 4.88MHz 8088 you shouldn't feel much performance penalty for audio DMA.
>-Hamid
AFAIK the NVIDIA has to reload it's texture information constantly. This
will take up a lot of bus bandwidth, not the samples.
Walter
dadai
> With a 66MHz 32-bit memory bus, reading 44kHz 16-bit samples (CD-quality)
> from memory will use up 0.032% of the memory's bandwith. I don't think
> anyone can feel a performance penalty of .00032, and even on a 33MHz 16-bit
> bus it is still imperceptable at just above .1%. Your argument about Video is
> true, but the bandwith of video is orders of magnitude above that of audio
> (640x480x16 at 60Hz requires nearly 37 million bytes per second, while
> CD-quality audio needs 88 thousand bytes per second, about 400 times less.
> Unless you have a 4.88MHz 8088 you shouldn't feel much performance
> penalty for audio DMA.
Sorry! Your caculation is wrong.
Suppose we have a MIDI files which we assign all 16 channels with
different instruments. Every instrument sample is 16bit, 44.1Khz,
mono (yes, most of MIDI sample in ROM is mono), so we got caculation
as following:
2 bytes (16bit) x 44,100 (sample point) x 16 (instruments) x 1 (mono)
= 1,411,200 bytes / 1024 = 1,378.125 KB/Sec
This figure show that the card has to burst this amount of instrumenst
sample data into its own memory. And the card itself has to has at
least this kind of memory to hold as buffers in order to cache theses
samples, otherwise you'll experience something maybe dynamic swapping
from card's memory and main memory to keep on loading samples. Will
Edge keep this amount of memory as memory buffers? It is
questionable.
Here we even don't include multisample which is a quite widely used in
MIDI synthesizer. Just imagine you will take a multisample of piano
for every octcave, you need 8 samples for piano only. One good
example is Ensoniq Soundscape, they only got 128GM instruments but in
ROM they store 512 seperate samples. What about multi-layer samples?
It will suck more bus bandwidth from you computer system..
Also, for PCI bus speed, the official figures from Intel is
133MByte/sec. It seemed quite a lot. But it is only theoritically
speed. If we look at CPU to DRAM speed, you can reference to PC
magazine's test. If my memory was right, I remeber PC magazine's test
is average 5MByte/Sec for 486DX2-66 and 10MByte/Sec for Pentium 66Mhz.
Pentium 90 or 100 should be able to reach 15MByte/Sec. Does PCI's bus
master function can be faster than CPU to DRAM speed? I doubt it.
I don't know how fast PCI's bus master can be, but for ISA bus, all
bus master function must use DMA transfer and it's 4Mhz or 4MByte/Sec.
I don't know how good Diamond's control (or chaching) software is.
But from previous post, it seemed Edge's sample has total 6MByte.
Than it might give you 2 options which is one to set 6MBytes memory
buffer in your main memory as sample cache or let you decide how much
memory you will set as your own requirement. First option you will
loss 6Mbyte of memory which can not be accessed by your application or
games and the second option, you'll experience HD swapping for your
instruments samples and it will further downgrade your system overall
performance.
Mac Chen
Tim Nielsen
>>having the capability of doing 2 billion ops/sec.
>>Pete
>>(guid...@tuns.ca 4th year CS)
>A followup to my last post. The thing is 9 square inches! Wow!
WHAT! No way. 9 inches? Let's see, the Pentiums are about 2" square, and they
hold like 3 billion transitors, right? So this thing has in the trillions? Not
likely. I read that it was 240 pins. If it was really 9" it would render it
useless. They claim it will be great to put on the motherboard. MY MOTHERBOARD
ITSELF is barely 9" square.
Are you sure about this?
Tim.
Peter Guidry
amazing. 3D,2D, audio, video, modem ... Check this out at www.mpact.com.
Frederick Y Mah
Sounds intereting. That thing must be huge.
--
Fred Mah --- fm...@ecn.purdue.edu
http://widget.ecn.purdue.edu/~fmah
Frederick Y Mah
>I don't know how this will work in practice, but on paper the MPACT chip is
>amazing. 3D,2D, audio, video, modem ... Check this out at www.mpact.com.
>From the specs I would bet this thing cooks. It uses rambus , and mentions
>having the capability of doing 2 billion ops/sec.
>Pete
>(guid...@tuns.ca 4th year CS)
A followup to my last post. The thing is 9 square inches! Wow!
Tim Nielsen
>I think you were thinking 9"X9" (which would be 81 in ^2)
Doh! I was kinda diggin the idea of a chip 9"x9" though, that'd be COOL! But,
I guess, 3x3 does make a little more sense. It's this whole 'square' thing, I
think I'm gonna have to bust out my old geometry thing.
Tim.
Dave Glue
Mah) wrote:
>Peter Guidry <stud...@bnr.ca> writes:
>
>>I don't know how this will work in practice, but on paper the MPACT chip is
>>amazing. 3D,2D, audio, video, modem ... Check this out at www.mpact.com.
>
>>From the specs I would bet this thing cooks. It uses rambus , and mentions
>>having the capability of doing 2 billion ops/sec.
>
>
>>Pete
>>(guid...@tuns.ca 4th year CS)
>
>A followup to my last post. The thing is 9 square inches! Wow!
Holy! Well, the only thing that matters is price, and apparently it's
going to be quite cheap. Certainly one of the most interesing
hardware developments of the year, 96 is going to be one hell of a
year for PC gaming and multimedia in general.
Dave Glue
Nielsen) wrote:
>
>WHAT! No way. 9 inches? Let's see, the Pentiums are about 2" square, and they
>hold like 3 billion transitors, right? So this thing has in the trillions? Not
>likely. I read that it was 240 pins. If it was really 9" it would render it
>useless. They claim it will be great to put on the motherboard. MY MOTHERBOARD
>ITSELF is barely 9" square.
>
>Are you sure about this?
>
Perhaps he meant 9 inches _squared_- meaning, 3" by 3". Dunno.
Dave Glue
Rohrmeier) wrote:
>Frederick Y Mah (fm...@whatsis.ecn.purdue.edu) wrote:
>: Peter Guidry <stud...@bnr.ca> writes:
>
>: >I don't know how this will work in practice, but on paper the MPACT chip is
>: >amazing. 3D,2D, audio, video, modem ... Check this out at www.mpact.com.
>
>: >From the specs I would bet this thing cooks. It uses rambus , and mentions
>: >having the capability of doing 2 billion ops/sec.
>
>2 Billion operations per second??? Wow! You mean it will do two billion
>No-Ops a second? Know how useless a no-op is?
>
>MIPS/BIPS are useless, meaningless, worthless, and highly over-used measures
>of performance. Stop using them.
Why not check out the web page, instead of shooting the messenger? I
greatly appreciate being kept up to date on these developments.
Regardless of his performance measure (he just listed some quick
specs, relax for godsakes!), the chip DOES look very impressive. Read
up on it.
2xw7ha...@vms.csd.mu.edu
>>>A followup to my last post. The thing is 9 square inches! Wow!
>
>>hold like 3 billion transitors, right? So this thing has in the trillions? Not
>>likely. I read that it was 240 pins. If it was really 9" it would render it
>>useless. They claim it will be great to put on the motherboard. MY MOTHERBOARD
>>ITSELF is barely 9" square.
>
>>Are you sure about this?
>
>
>Yeah. I looked at one of the images of the chip and the dimensions shown
>per side were 3.07 inches.
Quaeghebeur-Hoefmans
ERiK Q.
Hoang Pham
>
>WHAT! No way. 9 inches? Let's see, the Pentiums are about 2" square, and they
>hold like 3 billion transitors, right? So this thing has in the trillions? Not
>likely. I read that it was 240 pins. If it was really 9" it would render it
>useless. They claim it will be great to put on the motherboard. MY MOTHERBOARD
>ITSELF is barely 9" square.
The person says "9 square inches," not 9 inches squared. So, according
to him, the chip is likely 3"x3", which is still a fairly humongous piece
of silicon.
I checked the web site, and couldn't find the physical size listed
anywhere. I am somewhat skeptical of the "9 sq. inches" report, as the
company would have major problems trying to fit it onto anything less
than a full-length card, let alone finding real estate for it on a
typical baby-AT motherboard.
Paul Campbell
fm...@whatsis.ecn.purdue.edu (Frederick Y Mah) writes:
> A followup to my last post. The thing is 9 square inches! Wow!
read the web page again - that's the footprint of the chip and
other components (RAMDAC, dram, audio codecs etc).
It's NOT the die size.
---------------------------------------------------------------------
Paul Campbell - Taniwha Systems Design - Oakland CA USA
$cientology - the 'religion' for the '50s, where brainwashing is
a sacrament - if you think your newsgroup has wackos check out
alt.religion.scientology! (SP2 and bar)
Dave Glue
>I checked the web site, and couldn't find the physical size listed
>anywhere. I am somewhat skeptical of the "9 sq. inches" report, as the
>company would have major problems trying to fit it onto anything less
>than a full-length card, let alone finding real estate for it on a
>typical baby-AT motherboard.
>
You may see his post, as the mpact rep I just spoke to in e-mail
clarified things a bit, his news server is quite slow.
The 9" squared is the _board_, no the the chip. The board includes
the ram/dac, codec, all components necessary, etc. The chip is
certainly not 9" squared.
Frederick Y Mah
>He's telling the truth, the thing is 3" on a side. It also dissipates
>approx 180 watts (max, 20 watts when idle) and so you need a mechanical
(cut)
You are just kidding about the 180W right? The first P6 will need to have
cooling to handle about 20W and newer P6's up to about 38W.
That was, of course, my second thought "how much energy is it dissipating."
You could just slap on a Peltier cooler to it, that should work if it goes
to about 20-30W though you might need a larger PC power supply and extra
case fan.
Pascal Jasmin
-> WHAT! No way. 9 inches? Let's see, the Pentiums are about 2" square,
-> and they hold like 3 billion transitors, right? So this thing has in
-> the trillions? Not likely. I read that it was 240 pins. If it was
-> really 9" it would render it useless. They claim it will be great to
-> put on the motherboard. MY MOTHERBOARD ITSELF is barely 9" square.
9 square inches means 3"x3"
-----
I can't believe how stoopid bad farkin spellars everyone else is.
Frederick Y Mah
>In article <461vjb$p...@mozo.cc.purdue.edu>
>fm...@whatsis.ecn.purdue.edu (Frederick Y Mah) writes:
>> A followup to my last post. The thing is 9 square inches! Wow!
>read the web page again - that's the footprint of the chip and
>other components (RAMDAC, dram, audio codecs etc).
>It's NOT the die size.
>---------------------------------------------------------------------
>Paul Campbell - Taniwha Systems Design - Oakland CA USA
>$cientology - the 'religion' for the '50s, where brainwashing is
>a sacrament - if you think your newsgroup has wackos check out
>alt.religion.scientology! (SP2 and bar)
I guess the figure is wrong then? It shows a 240pin quad with dimensions
of 3.07 inches per each side.
Ian Buckner
: pa...@taniwha.com (Paul Campbell) writes:
I could not find the dimensions on the web site, however....
a 240 pin QFP usually has pin spacings of 0.5mm (NOT 0.020 inches, but that
is another story ;-). This would give a chip size just over 30mm square
(or about 1.2" square), which would be much more reasonable. A 3" square
chip would be a _major_ headache if it dissipated any significant amount
of power, as things like the thermal differential expansion would rapidly
fatigue solder joints.
Cheers
Ian Buckner
These opinions are mine, not hp's
sdaks
>In article <469kgc$e...@steel.interlog.com>, dav...@interlog.com says...
>what these new "chips" are. Are they add-on PCI cards with TONS of
>functions or are they new chips. I guess my question is whether or not I
>will be able to buy a Mpact or other DSP cards and plug it into my desktop
>computer at home.
Here's info from a news clipping. I just cut and pasted it here, so excuse the
margins. It's worth the hassle to readup... Also there's a WWW site out there
I think it's WWW.MPACT.COM... don't quote me on that though.
Single Chip Gives Home PC's HDTV-Quality Movies, Surround Sound, Videophone And Much More; Affordable Baseline of New Multimedia Capabilities Will Redefine 1996 Home PCs
MOUNTAIN VIEW, Calif.--(BUSINESS WIRE) via First! -- Chromatic Research, Inc., together with Toshiba America Electronic Components, Inc. and LG Semicon America, Inc. today jointly announced the Mpact(TM) media engine: the first complete, single-chip multimedia solution for the home PC.
The Mpact media engine provides state-of-the-art capabilities for all seven multimedia functions -- video, 2D graphics acceleration, 3D graphics acceleration, audio, FAX/modem, telephony and videophone -- in a single, software-upgradable processor designed for PC motherboards. PCs equipped with the Mpact media engine will be available to consumers in mid-96.
Mpact's audio capabilities include Dolby(R) Surround AC-3(TM), the audio standard for the recently announced digital video disk (DVD) format. Chromatic is the first computing company to license Dolby Surround AC-3.
Mpact also includes 3-D sound for games that fools the ear into thinking sounds are coming from beyond the points where the two speakers are located.
Mpact's video capabilities include MPEG-2 (Motion Pictures Experts Group) decoding for HDTV (high definition television)-quality movies and video clips. With the combination of MPEG decoding and Dolby Surround AC-3, users will be able to watch full-screen, feature-length DVD movies as well as instructional and entertainment CD-ROMs. These capabilities exceed the VCR-quality video of MPEG-1, which is just now becoming available in home PCs.
Mpact is the only product of any kind that meets the complete multimedia needs of the exploding $12 billion home PC market(1). For the same or even lower cost than today's limited PC multimedia capabilities, Mpact-equipped PCs will allow users to:
-- watch HDTV-quality video in movies, games and CD-ROMs;
-- play video games with realistic 3D imaging and 3D sound that rival the latest "64-bit" gaming systems from Sony, Sega and Nintendo;
-- experience realistic music and sound effects with better-than-CD-quality;
-- view and download video and graphics-intensive information quickly from the Internet and commercial on-line services;
-- view photo-realistic images with over 16 million colors;
-- talk to someone while sending them data over the same phone line;
-- use their PC as personal or family voicemail system with caller ID;
-- conduct business or keep in closer touch with friends through an affordable videophone.
The Mpact media engine consists of two components: 1) a single ultra-high-bandwidth, highly parallel media processor, and 2) "mediaware" software modules which enable the seven functions and various levels of capabilities within each function.
The Mpact media processor, architected by Chromatic, was a joint development effort by Chromatic and Toshiba for Toshiba's implementation of the Mpact media processor. A similar joint effort is underway with Chromatic and LG Semicon. Under license from Chromatic, Toshiba and LG Semicon will manufacture and sell their Mpact media processors.
Chromatic will sell its enabling mediaware modules directly to PC manufacturers.
Making PCs Even More Compelling For Consumers
The three companies expect the Mpact media engine to enable the affordable 1996 home PC to become the family entertainment, education and communications device. For example, an Mpact-enabled "PC" could also function as a set-top box for interactive television, a videophone, a high-fidelity stereo, a digital VCR and more. The Mpact media engine will make home PCs even more compelling to consumers while providing PC vendors with significant new market opportunities in the consumer electronics arena.
"The multimedia PC today resembles Frankenstein's monster, and it performs about as smoothly," said Wes Patterson, President and CEO of Chromatic Research. "Today's definition of a multimedia PC is one with a CD-ROM drive and a hodgepodge of separate add-in cards and various chips that work poorly with each other to offer limited capabilities. Mpact will establish a new feature-rich, high-performance baseline for PC multimedia, but it will go well beyond just making PC multimedia better. Mpact enab
les the PC to live up to its bold promise as the family's entertainment, education and communications center."
"I don't know of anyone trying to do what these companies are doing together," said Ted Waitt, President and CEO, Gateway 2000. "They are the only ones offering a complete multimedia solution in a single chip. The incredible list of multimedia functions that they bring to the PC platform should make the fast growing home PC market grow even faster."
The Mpact media engine is designed for the PC and is optimized to take full advantage of today's most widely used software standards and applications including Windows 95.
"We are excited about the Mpact media engine and its potential to show off the entertainment- and education-focused APIs (application programming interfaces) in Windows 95," said Jay Torborg, Director of Graphics and Multimedia in the Consumer Systems Division of Microsoft. "Mpact is designed to support a broad range of multimedia capabilities and will enable software developers to quickly capitalize on the advantages offered by Microsoft APIs such as DirectPlay, DirectDraw, Direct3D and upcoming Dire
ctVideo."
"We expect Mpact to find strong acceptance amongst system manufacturers for motherboard implementation," said Martin Reynolds, Director of Technology Assessment for Dataquest, Inc. "3-D graphics will be a key feature for Christmas of 1996. Mpact provides this feature, along with video processing, sound and telephony functions."
A New Business Model For Semiconductors
Chromatic Research will be the world's first "chipless" semiconductor company, due to a new business model that uniquely enables the company to meet home PC manufacturers' pricing and volume requirements for motherboard integration.
Toshiba Corporation, Tokyo, Japan and LG Semicon Co., Ltd. (formerly Goldstar Electron), Seoul, South Korea have licensed Chromatic Research's design and will manufacture and sell the Mpact media processor under the Mpact brand name. Both Toshiba and LG Semicon are gearing up for the production of millions of Mpact media processors in 1996 using the latest high-volume, low-cost semiconductor manufacturing processes.
This business model gives Mpact the cost and volume requirements needed to meet the demand of the entire consumer PC industry.
PC manufacturers will be able to integrate a fully-configured Mpact solution including memory and mediaware -- for under $150, about the same cost that PC makers currently pay for just a low-end 2D graphics card, basic 16-bit sound card and a FAX/modem.
Breakthrough in Technology
The Mpact media processor is a vector processor that combines Very Long Instruction Word (VLIW), the same approach taken by Intel and HP with the upcoming P7 processor, and Single Instruction Multiple Data (SIMD), an approach used in supercomputers, to provide the fastest multimedia performance ever offered.
The Mpact media processor is optimized for PC multimedia applications and is designed to work as a coprocessor to the main x86 CPU. The Mpact media processor does not include functions that can be effectively performed on the x86, such as operations involving floating point calculations (used in 3D graphics), but is dedicated to quickly and efficiently processing "natural data types" such as graphics, digital video and sound. While it would take at least 10 of the fastest Pentiums(TM) to do the job
of one Mpact media processor in accelerating multimedia, it would also take about 10 or more Mpact media processors to perform the general purpose functions of a Pentium.
One of the best measures of a chip's ability to handle multimedia tasks is the number of integer operations per second (IOPs) it can perform. For typical tasks, a 120 MHz Pentium performs about 200 million IOPs. The Intel P6 processor is expected to perform about 300 million IOPs. The Mpact media processor can perform 2 billion IOPs. Through a proprietary version of a technique called "motion estimation" that allows the processor to predict how video frames will change from one to the next, Mpact
can perform 20 billion IOPs for MPEG video encoding and videoconferencing.
Chromatic Research
Founded in 1993, Chromatic Research is redefining the PC by delivering the first and only complete multimedia solution. Chromatic's Mpact media engine consists of silicon technology and software that combines all seven multimedia functions -- video, 2D graphics, 3D graphics, audio, FAX/modem, telephony and videophone. The company is privately held.
Toshiba America Electronic Components
Toshiba America Electronic Components, Inc. (TAEC) is the North American engineering, manufacturing, sales and marketing arm of one of the world's largest suppliers of semiconductors, integrated circuits and electronic components for industrial and consumer applications.
The company is the recognized leader in CMOS technology and has one of the broadest IC product lines in the industry. In addition, Toshiba is a leading manufacturer of technologically advanced electron tubes and solid state devices, including color picture tubes, display monitor tubes, liquid crystal displays, medical tubes, rechargeable lithium ion batteries, microwave components, laser diodes and optical transmission devices. The ASSP unit is located at 1060 Rincon Circle, San Jose, CA 95131.
LG Semicon America, Inc.
LG Semicon America, Inc., the U.S. semiconductor subsidiary of LG Semicon Co., Ltd. and the $48 billion LG Group of South Korea, markets and distributes a varied line of integrated circuits, featuring memory, logic, and microperipheral products that are widely used in personal computers, printers, copiers, FAX machines, and a host of other industrial and consumer electronic products. The company's U.S. operations -- including its sales, marketing and administrative offices, its R&D and CAD facilitie
s, and its product warehouse -- are headquartered at 3003 North First Street, San Jose, CA 95034-2004.
Reader Contact Information
Readers who want more information about Mpact can call Chromatic Research 415/254-1600; Toshiba America Electronic Components at 408/456-8900, or call LG Semicon America at 408/432-5000. Readers can also visit the Mpact World Wide Web site at http://www.mpact.com (unveiled Oct. 9).
Note to Editors: Mpact is a trademark of Chromatic Research, Inc. All other trademarks and registered trademarks are the property of their respective owners.
(1) Link Resources, 1995. Figure represents 1994 home PC sales.
CONTACT: Chromatic Research | Susan Thomas, 415/254-0729; s...@chromatic.com | or | Copithorne & Bellows PR for Chromatic | Dave Wilt, 408/988-2100; da...@sv.cbpr.com | or | Toshiba | Jim Lucas, 408/526-2535; jnl...@taec.com | or | LG Semicon | Arun Kamat, 408/432-5000; ar...@lgsa.com | or | Shafer PR for Toshiba | Pam Sloane, 415/989-0595 x 112
[10-09-95 at 07:20 EDT, Business Wire, File: b1009071.700]
Entire contents (C) 1995 by INDIVIDUAL, Inc., 8 New England Executive Park West, Burlington, MA 01803.
Tim Kraemer
>Single Chip Gives Home PC's HDTV-Quality Movies, Surround >Sound, Videophone And Much More; Affordable Baseline of New >Multimedia =
Capabilities Will Redefine 1996 Home PCs
>The Mpact media engine provides state-of-the-art capabilities >for all seven multimedia functions -- video, 2D graphics >accelerati=
on, 3D graphics acceleration, audio, FAX/modem, >telephony and videophone -- in a single, software-upgradable >processor designed f=
or PC motherboards.
PLEASE tell me that you dont really believe that this card
will really be able to do all these things reliable, or
that it will even be able to actually do them all.
Dave Glue
Daniello) wrote:
>The catch is each MPACT card would be able to handle a couple of
>those functions, not all of them. You might have a MPACT
>video card that does MPEG stuff for example. For audio you'd
>need a separate MPACT based card. It still looks good though
>since they purport the chip will be cheap.
The Mpact is _one_ processor, that can handle _all_ of those
functions. Now, you would probably need an external codec for
sound/modem/telephony, but not much more. You could even use the
Rambus dram for sound if you wanted to. Software "modules" can be
downloaded into the Mpact at will to support different functions- most
of the functions overlap the external components necessary.
Floyd D. Whitehurst
Jakob Brundin
In article <4795qj$5...@oznet07.ozemail.com.au>, r.f...@student.anu.edu.au (Richard Ford) writes:
>dav...@interlog.com (Dave Glue) wrote:
>
>It will sound crap! The more cicuitry and crap the sound has to
>travel trough and around, the more loss of the audio qualitry you get.
>
Does the sound travel like analog signals or is the chip a DSP?
Only after the DA conversion the sound is sensitive to noise.
>Why do you think high quality sound systems are split into power
>amp/pre amp tuner and all the other seperate units? How come you
>never see the "wonder amp', with cd and coffee maker built in?
>
Don't mix power supplies (producers) with audio circuitry (consumers),
there is quite a difference.
Jakob
Matt Masuda
: In <4744ks$l...@steel.interlog.com>, dav...@interlog.com (Dave Glue) writes:
: > The Mpact is _one_ processor, that can handle _all_ of those
: >functions. Now, you would probably need an external codec for
: >sound/modem/telephony, but not much more. You could even use the
: >Rambus dram for sound if you wanted to. Software "modules" can be
: >downloaded into the Mpact at will to support different functions- most
: >of the functions overlap the external components necessary.
: I didn't say there was more than one MPACT processor, I said you'd
: need several of them to be able to use all of those functions
: at once. The MPACT home page even lists typical functional
: combinations you might see in a card. Read the platforms page on
: www.mpact.com/Products/platforms.html. It suggests combinations
: including: an audio + fax/modem board, a video + 2D and 3D
: acceleration. a video card with TV capability, and an
: MPEG encoder + 2D/3D graphics. They do *NOT* suggest that
: with one MPACT card you'll be able to do all of that stuff.
They don't say you can, but neither do they say you can't. Check out the
datasheet at http://www.mpact.com/Technical/specs.html#ds01.
The Mpact processor uses "mediaware modules" to enable each of its
features. The chip runs a multitasking kernel which runs multiple
modules concurrently. There is no mention of a limit on how many modules
can be running at one time. I'd guess it's just a matter of having the
proper I/O hardware and loading up all the modules.
--
* Matthew C. Masuda * "Darkness rules!" *
* mcma...@greatbasin.com * - Beavis *
dennis r. gorrie
version of MK3 vs the Playstation version, (using a TV) , would suffer due
to being exposed because you sit closer to the computer monitor.
HUH?
First of all, when I see kids playing video games on a TV, where are they
sitting? RIGHT BY THE TV! And out of curiosity who do you think has the
slacker FCC regs wrt E-M rad? Consumer televisions have the lower regs
to adhere to, in many cases comparable regulations don't even exist! I'm
wondering if the original poster has ever used an E-M measurement tool like
those by Tek, Fluke, etc? I don't think so... if he had he wouldn't be
standing there with that size 13 foot in his mouth. (Right next to that
glowing, festering, head wound).
(FYI, most computer monitors peak radiation levels are a fraction of peak
levels I've seen from consumer televisions)
As a 30hr/week PC gamer with a 10cmx3cm tumor in my frontal lobe, I
believe this information to be true, accurate, and un-biased in any way!`
G A M E O V E R
------------------
;-) <--- (aid for the humor impaired) ---> (-;
Rolande Kendal
r.f...@student.anu.edu.au (Richard Ford) wrote:
>dav...@interlog.com (Dave Glue) wrote:
>
>>On Mon, 30 Oct 1995 21:42:47 GMT, dani...@austin.ibm.com (Christopher
>>Daniello) wrote:
>
>
>>>The catch is each MPACT card would be able to handle a couple of
>>>those functions, not all of them. You might have a MPACT
>>>video card that does MPEG stuff for example. For audio you'd
>>>need a separate MPACT based card. It still looks good though
>>>since they purport the chip will be cheap.
>
>>functions. Now, you would probably need an external codec for
>>sound/modem/telephony, but not much more. You could even use the
>>Rambus dram for sound if you wanted to. Software "modules" can be
>>downloaded into the Mpact at will to support different functions- most
>>of the functions overlap the external components necessary.
>
>travel trough and around, the more loss of the audio qualitry you get.
>
>amp/pre amp tuner and all the other seperate units? How come you
>never see the "wonder amp', with cd and coffee maker built in?
>
>
>
This person does not seem to understand the concept of digital technology.
Rolande < The sun shines on everyone! >
Mike J. Scott
dennis r. gorrie <vd...@orion.calgary.chevron.com> wrote:
>The original post of this thread, suggested that users of the Computer
>version of MK3 vs the Playstation version, (using a TV) , would suffer due
>to being exposed because you sit closer to the computer monitor.
>
>HUH?
>
>First of all, when I see kids playing video games on a TV, where are they
>sitting? RIGHT BY THE TV! And out of curiosity who do you think has the
>slacker FCC regs wrt E-M rad? Consumer televisions have the lower regs
>to adhere to, in many cases comparable regulations don't even exist! I'm
>wondering if the original poster has ever used an E-M measurement tool like
>those by Tek, Fluke, etc? I don't think so... if he had he wouldn't be
>standing there with that size 13 foot in his mouth. (Right next to that
>glowing, festering, head wound).
>
>(FYI, most computer monitors peak radiation levels are a fraction of peak
>levels I've seen from consumer televisions)
I agree completely. You're exposed to A LOT more EM radiation from your
hair dryer or the train commute into work than you'll receive from your
computer monitor. From the csiph.video FAQ:
Q 5.16) Should I be concerned about monitor emissions?
[From: Michael Scott (sc...@bme.ri.ccf.org) and Bill Nott
(BN...@bangate.compaq.com)]
CRT's (Cathode Ray Tubes) direct a beam of electrons at a thin layer of
phosphor which coats the screen on your monitor. When the electrons
strike the phosphor, shadow mask and other screen components, x-rays are
produced. The amount and energy of the x-rays depends on the
accelerating voltage. The relatively low voltages in CRT's (compared to
commercial x-ray machines) means that relatively low quantities of low
energy x-rays are produced and modern monitors are so well shielded, that
there is no concern of being irradiated over time. Though it is possible
for a damaged monitor to emit x-ray radiation, it is unlikely that
harmful amounts will be released, and most x-rays would be directed
towards the back or sides of the monitor. Any damage to the front of
the CRT severe enough to increase x-ray emission would cause the CRT to
implode.
All televisions and computer monitors must comply with various worldwide
standards for ionizing emissions. Information relating to this compliance
is typically included on the product label, or within the users manual.
Recently, concerns about low frequency (LF) and very low frequency (VLF)
electro-magnetic and electro-static emissions have been raised. Many
studies have been established recently to determine if these concerns
are warranted. None of the studies has concluded that there is any
correlation between the radiation and possible health risks. In Sweden,
a large study was undertaken and as a result, the Swedish government,
and the Swedish Workers Union (TCO) both established recommended limits
of radiation for office equipment, including Video Display Terminals
(VDT's). The same limits are applied to monitors; the Swedish Government
standard is referred to as MPR 1992, and the TCO standard is referred
to as TCO. Many new monitors adhere to the Swedish emission regulations.
Epidemiologists have suggested that the risk factors for some childhood
cancers (particularly leukemia) are as high as two for some populations
exposed to low frequency EMI. A risk factor of two means that the odds
of being afflicted with a disease is twice as likely in the exposed
population than in a control population. In general, a risk factor of
less than six is not considered significant (cigarette smoking has a risk
factor of 10-20). As a result, several groups have publicly stated that
there is no significant health risk from EMI radiation levels experienced
by people from home appliances or nearby high voltage lines.
Critics of the Swedish study suggest that it was simply too huge.
According to a television documentary, over 800 comparisons were made
for correlation between exposure and pathologies. Statistics would
suggest that given enough completely random and uncorrelated measures,
the odds are that some of them will display a high correlation. As a
result, any study that is large enough will produce correlations
between _some_ of the measured quantities. Because of this criticism,
and the fact that only correlation, and no causation was proved in the
study, the Swedish government has since reversed their decision to
mandate maximum EMI emissions.
Studies in the U.S. to determine if EMI could cause cancer or other
illness, birth defects or any other health problems in rats have
come up negative. The rats were exposed to 0-10000 mG (milli-Gauss)
magnetic fields (the earth's magnetic field is ~500 mG), and their
skeletal and visceral organs, reproduction, frequency of cancer and
immunology all came up normal. The chronic studies that were
undertaken by the same group will be completed in 1996. Other studies
showed that EMI had no effect on the growth of cancer cells.
So, you have to make your own decisions, but the overwhelming majority
of experts agree that there is no cause for concern.
ciao,
--
Michael J. Scott R.R.I., U of Western Ontario
mjs...@heartlab.rri.uwo.ca 'Need a good valve job?'
############### Illegitimus non tatum carborundum. ##############