Slashdot Mirror
Matrox Parhelia 512 Preview
SpinnerBait writes "Finally, you don't have to sift through all the unreleased and unauthorized
bogus information around the net about Matrox's upcoming 3D Graphics chip,
called the Parhelia 512. Matrox has taken the wraps off their next
generation GPU and
this Preview over at HotHardware goes through its feature set with a fine
toothed comb.
They also give you a very rare glimpse inside Matrox's Montreal Headquarters,
as well as a look at some very impressive technology demos, rendered on their
new chip. Looks like impressive stuff for sure."
This Graphic card as almost more processing power then my two PC's combined! The only thing I wish is that Matrox could come back a bit in this market. They had made some good card in the past. More choices can only be good. And, do they have a good record of supporting Linux in the past? Funny, they are located in my town and I know less about them then all those US based company :)
I'd rather be sailing...
Kyle over at [H]ard|OCP has quite a load of info on the card:
h tm l
i s. html
http://www.hardocp.com/articles/parhelia/index.
http://www.hardocp.com/articles/parhelia/analys
...to release decent drivers. Tested and stable would be nice...
Good job my pc Blew up the other day, I now have a great excuse to upgrade..
Or more seriously, I wanted to upgrade before (from my G400) but GForce / ATI have poor 2D performance and some bad filters on there cards which require a bit of hacking to sortish out, and Matrox didn't have a viable home/gamer solution, sure there 10bit medical cards look nice, but not quite for me.
The only problems i have had in the past with matrox cards are,
Poor OpenGL support, though the drivers seemed to have been fixed as of Feb this year.
There Linux support is a little, well patchy. they do provide drivers, but there only half open and a bit of a pain to get working corretly, some of the problems may have been down to old X4 versions though.
Well I'll Buy one in the next couple of months and try to post a more informed comment!!
thank God the internet isn't a human right.
Here are some more links with Parhelia info:
h tm l
w z? cid=3&aid=425
h om e.cfm
http://www.hardocp.com/articles/parhelia/index.
http://www.hardwarezone.com/articles/articles.h
http://www.matrox.com/mga/products/parhelia512/
I hope they make Linux drivers for it. Hardware text AA seems kinda cool.
Here
The summary mentions quite a few interesting notes regarding the effect this card would have on current games.
- In "simple" games like Quake III Arena, the Parhelia-512 will definitely lose out to the GeForce4 Ti 4600. By simple we mean games that generally use no more than two textures and are currently bound by fill rate. NVIDIA's drivers are highly optimized (much more so than Matrox's) and in situations where the majority of the Parhelia's execution power is going unused, it will lose out to the Ti 4600. This can change by turning on anisotropic filtering and antialiasing however, where the balance will begin to tilt in favor of the Parhelia.
- In stressful DX8 games, Matrox expects the Parhelia-512 to take the gold - either performing on par or outperforming the GeForce4 Ti 4600. Once again, as soon as you enable better texture filtering algorithms and antialiasing the Parhelia-512 should begin to seriously separate itself from the Ti 4600. The quad-texturing capabilities of the core as well as the 5-stage pixel shaders will be very handy in games coming out over the next several months.
So from the look of it, Parhelia does not wipe out Nvidia (though I would like them to), but is a worthy competitor to nvidia in current games. It would be interesting to see how ATI and Nvidia match up to this new competitor in the coming months.
Be afraid. Be vewy vewy afraid.
Rapid Nirvana
Well from reading a few artivles about needing more than 8bit per channel, its all down to bleading.
It's a bit like using 24bit sound recordings to mix and then downsampling them to 16bit.
thank God the internet isn't a human right.
I wonder if their new Parhelia can deliver on its promises? Have Matrox's openGL drivers improved significantly over the past few years? Poor openGL was what killed G200's promising future, and I would hate to see a repeat performance.
24 bit (16 million colors) are a lot, and I certainly have difficulties to find the difference between color #70e0e0 and #70e1e0, but when you want to have a nice background, top is plain blue (#0000ff) and botton is black (#000000), then there are only 254 levels between those. And I can clearly see those lines where the blue color value changes.
And that's where the more colors shine. Just using 10 instead of 8 bits reduces those color bands by a factor of 4.
Instead of not using those alpha bits at all (in 32 bit color mode), one might as well use them for nicer colors. Now which OS supports that mode? X11?
HaraldWell, it looks like Matrox may be back into the mainstream. To most consumers, they're an unknown. To techies, they're the little company that refused to die, and to businesses, they're the best supplier. We'll see which of those three items changes.
Michael C. Hollinger
I hope it's better than their sorry excuse for a TV card that never worked right. It was a few years old, but it was an add-on board to one of their video cards(VFW), and the framerate changed wildly, if it worked at all.
Ok.. so it has AGP 8x. Nifty! What motherboard do I buy that has AGP 8x? I just bought an Abit KR7A-RAID with Via KT266 chipset, thinking this is a pretty decent board, but I doubt it supports AGP 8x.
Now we move on to monitors. Could someone recommend a monitor that I can use to accurately resolve 1 billion colors? I tend to run my 2 Viewsonic PT775's at 1600 x 1200 so I've grown accustomed to that much "real estate".
This sounds like an awesome card, but I really don't know where to go or what to get to reap all the benefits of it.
Lastly, precisely when and where can a fellow technogeek acquire one? Since the HotHardware site seems to be experiencing some serious "Slashdot Effect" I was unable to finish reading the entire article. MRP $$ and a release date would be very useful.
Vortran out
Knowledge is like ignorance.. too much can be just as bad as not enough.
The human eye can distinguish about 10 million different colours. But it's more sensitive to some frequencies than others, so sometimes 24 bits (16 million colours) may not be enough.
For example, most people can distinguish between two very similar 24-bit medium greens but not between three or four similar 24-bit dark blues.
That said, no monitor can accurately represent 16 million colours, let alone several billions. Even if they could, the dynamic range of monitors is very limited compared to the range our eyes can see (ie, monitors have very limited brightness compared to the normal sunlit world), so most of those colours would be wasted.
Higher colour precision is good because it minimises round-off errors, but this applies mainly to internal calculations (some operations are done directly on the final framebuffer, but very few). For display, 24 bits (and a good monitor) are more than enough.
RMN
~~~
Worried execs decided to announce the launch of the GeForce 5 later this year.
I kid you not!!!
I'm currently running an AGP Matrox G450 with 32mb of RAM with two CRTs. I like the card because it allows me to go up to 3200 x 1200 resolution with 32bit color.
...
I really like the prospect of having three monitors to eliviate the issue of having a giant gap between displays due to the thick boarder of any display. However
This new card claims it only does 3840 x 1024 resolution on three cards. It still has the max color depth, but the resolution has to drop. By going to this big fancy new card I'd only gain 100,000 pixels, which in reality is next to nothing.
Is it a driver limitation, or does it take more than a 512bit dual 400mhz 256mb video card to push 4800 x 1200 for simple 2D functions?
~LoudMusic
No sig for you. YOU GET NO SIG!
This is the only picture I could find of Parhelia.
Look at the massive heatsink on that baby... Ooooh mama...
Rapid Nirvana
For those of you who don't already know, professional TV standards (specifically, D1, also known as SDI, though SDI is technically different) use 10-bit YCrCb video.
This means that any particular pixel may have up to 30 bits of color (even though the maximum difference between colors of pixels is less than that.
Obviously, this is not something that is easily accomplished with standard 24 bit/32 bit rendering. If you convert the SDI into something that can be represented in the frame buffer of the video card, then you've lost precision. This is unacceptable for broadcast! (And no, overlay isn't generally good enough since you want to capture the pixels for output though SDI)
Admittedly, this card isn't perfect- It would be nice to have 8 bits of destination alpha (for a key channel). 4 shades of keying just isn't enough...
In any case, having a card (finally!) support 10 bit rendering (especially the 10 bit rendering in openGL) in hardware will be wonderful!
Keep in mind that there is a difference in the frequency response of your monitor for luma and chroma!
We are much more attuned (visually of course) to difference in intensity (luminance) in comparison to difference in color. IT is possible for monitors to have a higher luma response than chroma response, so...
Having 30 bits worth of intensity variation (ok, call it 10 bits if you will) can provide more distinct rendering, wven on monitors that don't support 1 billion "colors".
Just when I thought that my workplace would never spring for a card with these features, up popped Page 6 (just ignore all those pictures of people playing games with the card) with Glyph Antialiasing for "business appeal!" Three monitors, here I come.
Have fun: Join D.N.A. (National Dyslexics Association)
Hopefully Matrox will discontinue the DualHead, TripleHead, etc., naming conventions before they get to the sixth generation (for the same reason that Intel didn't release a Sextium).
Have fun: Join D.N.A. (National Dyslexics Association)
Take a look at this explanation which explains what a parhelia is =)
interesting stuff
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
The G200 delivered beautifully on everything it promised. It allowed me to run 4 monitors. It wasn't a gamers chip, it was intended to help show more info than previously possible. Using the PCI version in Win2K with a special patch, I saw one PC that had 16 monitors attached. Amazing.
Do you want to remove linux?
This card looks really sweet, and Linux could really use some competition to NVIDIA in the 3d card market, I hope the Linux drivers are up to par.
If they're binary only, I hope they put as much effort into them as NVIDIA does.
Sticking feathers up your butt does not make you a chicken - Tyler Durden
How does having a third monitor eliminate the problem you describe(gaps between monitors)? If ANYTHING it would just make things worse, not better - that is unless you also switched to borderless lcd panels. or removed the casings from your crts.
Because the gaps aren't right in the middle of your field of view. With three displays, default pop-up behavior (dialog boxes and such) would occur on a single monitor in the middle of your viewing window without being split in half (unless it's one of those silly dialogs that's shown as a percentage width of the desktop).
There are, of course, ways to modify that behavior (using Matrox's own tools if you have one of their cards), but it would be nice to look straight ahead and get an uninterrupted center desktop. As it is now, I look straight ahead and see 3 inches of border and gap between desktops.
I spend all day now looking either to the left or to the right. Looking to the middle would be a nice change.
And the quip about reading the article was unnecessary. First, he could have tried to read the article and found the site suddenly unresponsive (as did I). Second, he could have been reading it for technical information instead of marketing info. It happens.
What about in 3D? I would think going through DRI to the hardware and not through X you'd be able to use the gigacolor functionality. And that's where it would be most handy - I don't really care about billion-color icons on my desktop.
Ooh, on second thought, even 24-bit color doesn't do well on the gradients in title bars and background images sometimes....
Hey kids, there's only 5 days left 'til Yak Shaving Day!
[...] but when you want to have a nice background, top is plain blue (#0000ff) and botton is black (#000000), then there are only 254 levels between those. And I can clearly see those lines where the blue color value changes.
Actually, you should make that black to blue to white. And while you'll manage to distinguish the colours at the centre of the scale (near "pure" blue), I doubt you'll be able to distinguish the colours at the top and bottom (near white and black).
The limitations of 24-bit colour can also be dealt with with dithering. Most high-end animation programs render internally at 48 / 64 bits per pixel (16 bits per component) and then dither the image when they convert it to 24-bpp (8-bpc). This would result in a much smoother transition from black to blue (and then to white), with no visible banding.
Most modern graphics cards already do real-time dithering, but only in 16-bit modes (which still work internally at 24 / 32).
RMN
~~~
Matrox has a history of abandoning large sections of their users. They left owners of the Motion JPEG hardware high and dry when they decided that it was too difficult to get the hardware working correctly, and that it was better to run it without the hardware acceleration. Those who had spent hundreds for hardware-accelerated video recording were left with a system that was comparable to ones available for $30 or $40.
The Tech Report with their in depth preview http://www.tech-report.com/etc/2002q2/parhelia/ind ex.x?pg=1
- First off, Q3A is used as THE single standard metric to see how a card will perform under a common load. It's a very good way to judge the raw speed of a card overall, and often provides good pointers as to overall performance in fancier modes or other games, but it certainly doesn't mean every game you play will be 100+ fps.
- Second, that figure is an AVERAGE. When actually gaming, the average framerate is not the issue - the MINIMUM framerate is the killer. 60 fps average is fine, but when the framerate drops to 10-15 fps in a heavy firefight, you're in trouble. A higher average framerate usually translates to a higher minimum as well. In fact, many sites have taken to quoting minimums as well, or even showing a complete framerate graph.
- Third, the ability to manage 100 fps at e.g. 1024x768 means only around 40 fps at 1600x1200, if your monitor extends that far, or perhaps only 30 fps at 1024x768 with 4x AA if it doesn't. Your card will need to score 200 fps if you want to improve your resolution/AA, or maybe even 300 fps if you want to do that and still keep your minimum fps above 60.
- Fourth, the same argument applies to other quality improvements like trilinear and anisotropic filtering. Taking 32 texture samples instead of 4 can really kill your framerate, so you better hope you're getting enormous framerates with non-anisotropic filtering if you hope to get acceptable speed with anisotropic filtering enabled.
- Fifth, Q3A is not the only game out there. There are a lot of more demanding games available today, even those based on the Q3A engine like RtCW, that will give you much lower framerates.
Combining two or more of the above factors can bring the fastest graphics card to its knees, even if it scores 200 fps in Q3A. We'll have to wait until we see scores of 300 or 400 before we can expect to play Jedi Knight II at 1600x1200 with 9x AA and 16-sample anisotropic filtering, while never dropping below at least 30 fps. But boy, will it look good when we can :-)
Ideally, a review will give individual scores for all the above - high resolution, AA, anisotropic filtering, a range of modern games, and all combinations of the above. But since this would entail a vast amount of testing and a huge array of numbers, most reviews settle for a few known tests that are indicative of performance in other tests. And the most popular of those is good old Q3A.
Why would anyone engrave "Elbereth"?
Matrox doesn't actually have a good history of getting cards out in a decent time frame. Figure that by the time this card is actually available (anyone remember the g400? how many months did it take to get one after it supposedly became available?) it will be irrelevant.
The next problem is that Matrox ruined their reputation in my eyes with the G200 by lieing about OpenGL. Lieing about how they were going to have it in November, then December, and so on... they kept this up until they announced the G400 and then suddenly the g200 was a no-go.
Ever since the G400 series it seems Matrox has been coming up with feature laden cards... trouble was no one asked for the features they chose to offer. Now they added even more features and a buttload of performance to boot. Yet as before, GF5 will be announced about the time this card is supposed to ship, and most likely be in stores at the same time.
* Winners compare their achievements to their goals, losers compare theirs to that of others.
Did you actually bother to read a manpage?
Your wrong.
'man X', under 'COLOR NAMES'
The syntax is an initial sharp sign character followed by a numeric specification, in one of the following formats:
#RGB (4 bits each)
#RRGGBB (8 bits each)
#RRRGGGBBB (12 bits each)
#RRRRGGGGBBBB (16 bits each)
As someone else pointed out, 10 bits of RGB does not equate to 10 bits of YUV. The Parhelia will give great 10 bit RGB previews (completely independant of output quality), and will even output a 10 bit YUV video signal - but only via S-Video, where the two colour signals get encoded together anyway. You need 10 bit component output, or 10 bit SDI, neither of which can be done by the Parhelia. It's more aimed at the 10 bit DVD market than a professional output solution.
The two-bit alpha limitation is largely irrelevant. For display on a monitor, RGB is all you need. Processing of deep-colour images should be done with at least 16 bits per component (including alpha) in memory for best results, then dithered down to 10 bit RGB for display. Key channel output requires a second video connector, so it won't do that at all.
Why would anyone engrave "Elbereth"?
I wasn't able to find any info from the hothardware or matrox sites. Any rumors as to when this is coming out, and how much it's going to cost?
Everything on HOCP is available from Matrox directly here.
Why would anyone engrave "Elbereth"?
Which means, if you want to run all monitors at the same res (required for "Surround Gaming", really), you're limited to the resolution of the external DAC, which probably struggles to do 1280 x 1024.
It's nothing to do with the driver, and you can always add a second PCI gfx card for more monitors to get all the area you need. Try 5 x nVidia Quadro4 400NVS cards, each with 4 monitor outputs capable of 2048 x 1536, for a total of 61 million pixels - 16 times what you have now :-)
Why would anyone engrave "Elbereth"?
Their edge-AA functionality would lend itself well to font rendering. It's debatable whether it'll help the speed or even quality of current Windows font rendering, but so long as you're not forced to use it, it can't hurt. The hardware gamma correction is good, and it does "de-gamma" the background before blending in the text (which should be done with linear data).
My question is, does it correctly support hinting? It's not much use unless it does.
Why would anyone engrave "Elbereth"?
Think I'll talk to my contact at Matrox to see if we can get ahold of one of these and support this mode.
Why would anyone engrave "Elbereth"?
That can not be correct. The XColor structure, which is used all over the Xlib API for communicating color values is 16 bits per gun, 48 bits total.
Try thinking that bit of logic over again. :)
Then try this experiment. Open a paint program, select a nice medium, fully saturated blue and paint 1/2 the screen with it. Now edit that color and change one of the color components by a single value. I.e. from FF to FE or something. Then paint the other half of the screen.
MOST people can see a mach band where the two colors that differ by only 1 value of one color component meet in the middle. (If you don't see it at first, try a slightly darker shade of blue) That is why you need more than 256 values per component. Even when you are only showing two of the 2^24 colors on the screen, the fact that there are "only" 2^24 colors becomes a limiting factor.
10 bits (1024 values) is reaching the level of human perception for the most part. BUT that's still not good enough because gamma correction in the hardware can reduce the actual color resolution back down to 8 bits pretty fast. Eventually we'll all be using 16 bits per component all the way through. (Well for graphics work anyway) That'll give enough user color matching and adjustment and hardware color matching enough "breathing room".
Contrary to popular belief, coding is not all free blow-jobs and beer. Those things cost MONEY!
That Matrox has all sorts of nice pictures to show off their 10-bit technology? But when you view it with your own video card, the most you'll get is 8-bit color. So what's the point of all the pretty pictures? Talk about the marketing folks not getting the point!
Software sucks. Open Source sucks less.
I demand Nerdity points for counting down last night the hours (minutes) until the NDA ran out.
:)
And sending all of my friends half-hourly updates.
Unfortunatly I did have to go to bed and was thus unable to be the first non-NDA'd person to read the previews.
Product name too hard to spell, help!
I will just keep on refering to it as the G1000, SOOO much easier, heh.
Need help treating your acne? Come here!
Extremetech checked out 3Dlabs offer instead:9 6,s=1017&a =26271,00.asp
7 &a =26865,00.asp
http://www.extremetech.com/article/0,33
Which in my eyes sounded a lot better than Matrox offer since it was much more general-purpose. But on the other hand Matrox knows what features are really needed, and the PS2 showed that general-purpose features won't get you anywhere if they are hard to use. Featurewise it's a draw, but they are two different kind of beasts.
Extremetech also has a thorough discusson of the Matrix release:
http://www.extremetech.com/article/0,3396,s=101
And don't blame me if that site don't have persistant links.
http://www.matrox.com/mga/start/newsletter/may_200 2/parhelia512.cfm
Nope. Cones are sensitive to red, green and blue (hence the use of RGB in TVs). [...] Luminance really isn't that important.
I recommend reading a bit more on the subject before making such definitive statements. You can start with this:
Spectral sensitivity of the human eye
As you can see, at 650 nm (pure red), the cones are almost blind. The brain combines this information with what it gets from the rods (luminance) and realises that there is some colour there. And since it has no blue, almost no green and only a little yellow, it's translated to "red".
TVs use RGB (red,green,blue) just as they could use CMY (cyan,magenta,yellow) or any other group of complementary colours (of which there is an infinite number - any three colours that are 120 apart in a spectrum wheel will do). It has nothing to do with the actual wavelengths that the receptors in our eyes are tuned to.
You may also want to read some more about how TV colour signals are encoded (messy but interesting) and why current standards are as they are. Do a quick search on the internet and I'm sure you'll find plenty of pages about it.
RMN
~~~
I think his point was that we evolved this way because blue (and red) are less "important" than green in the natural world. Which might be the case, who knows? Personally I think a lot of things that can be "explained" by evolution are more likely the product of chance.
Things like earlobes, pubic hair, and the fact that Windows 2000 is actually quite stable.
RMN
~~~
I only see two monitor ports on the back of that puppy...(and oh the agony of knowing I already have a G450 running dual under Linux....)
The screen caps make me drool
What is your Slash Rating?
Matrox has provided some really good support in the form of driver software and forums for various flavors of Windows and Linux. I believe there is also some level of FreeBSD support. I would be surprised if this one didn't have non-Windows support. I can't speak for non-x86 support.
What is your Slash Rating?
Evolution IS chance, it's just that the ones lucky enough to get the good changes breed more.
:)
:-)
No. Evolution means improvement. The theory of evolution through natural selection (which is what most people mean when they say "evolution") says that species tend to improve naturally when those improvements increase their probability of reproducing successfully.
But it's not always clear if a certain change will improve a species chances of reproduction and / or survival.
And when a certain characteristic has little or no relevance in survival and reproduction, then it will stay or go based purely on chance.
The reason we are more sensitive to green is because there was something related to it that allowed our ancient ancestors to get laid more
And if you can find out what this was, your theory may be right. Personally I cannot. That's what I meant when I said that there's a certain tendency to use the theory of evolution through natural selection to "explain" things that do not fit its definition.
If this were not true, then there would be more people who were not more sensitive to it.
The reason why we are more sensitive to green is a natural consequence of two things: 1. green wavelengths are at the middle of our visible spectrum and 2. our photon receptors aren't 100% accurate, so they don't react to just one wavelength. The result is that the blue and yellow receptors are also partially sensitive to green, so there's an increase in green "resolution".
It's sort of the way single-CCD cameras work (two in each four pixels is green). This translates not only to better spatial resolution but also to better colour accuracy.
One of these likely advantages would be the ability to more easily distinguish between plants that would kill you, and those that are nutritious.
You can't distinguish between poisonous and edible plants based on colour. There are poisonous and edible plants of just about every colour and shade.
Most animals are colour-blind and are quite able to distinguish what they can eat from what they cannot. Smell (and experience) are much more important than colour.
In fact, it works the other way around. Since almost all insects can see colour, it plays a major role in plant reproduction and survival, because the plants with the most striking colours will attract more insects and therefore reproduce more.
Human vision could be improved by covering a slightly wider spectrum, but there's no "natural" incentive for that to happen, so it doesn't. Women won't magically fall in love with me and ask me to be the father of their children just because I can see ultra-violet light (er... will they?).
When something is good enough, it'll stay that way for a long time. Nature is lazy. Which makes me a naturist.
RMN
~~~